[ { "slug": "microsoft", "name": "Microsoft", "url": "https://ai.engineer/orgs/microsoft", "summary": "Microsoft-affiliated AIE appearances connect enterprise retrieval with practical agent development: Pablo Castro covers Azure AI Search, Cedric Vidal demonstrates agent evaluation, and Marlene Mhangami examines testing AI-generated code. Across the recordings, workshops and technical demonstrations address how applications obtain context, expose tools, and make model-driven behavior testable, observable, and subject to application controls.", "speakerSlugs": [ "sharmila-chokalingam", "cedric-vidal", "yohan-lasorsa", "nagkumar-arkalgud", "keiji-kanazawa", "marlene-mhangami", "david-smith", "miguel-martinez", "michael-albada", "liam-hampton", "julia-kasper", "den-delimarsky", "hanchi-wang", "ornella-bahidika", "emma-ning", "chris-noring", "harald-kirschner", "lachlan-ainley", "amy-boyd", "nitya-narasimhan", "pablo-castro", "daniel-rosenwasser", "pamela-fox", "gabriela-de-queiroz", "aishwarya-srinivasan", "tisha-chawla", "susheem-koul" ], "talkSlugs": [ "accelerate-your-ai-journey-with-azure-ai-model-catalog", "agentic-excellence-mastering-evaluation-of-ai-agents-with-azure-ai-evaluation-sdk", "ai-didn-t-kill-the-web-it-moved-in-olivier-leplus-aws-yohan-lasorsa-microsoft", "ai-red-teaming-agent-azure-ai-foundry-nagkumar-arkalgud-keiji-kanazawa-microsoft", "beyond-code-coverage-functionality-testing-with-playwright", "build-evaluate-and-deploy-a-rag-based-retail-copilot-with-azure-ai", "building-applications-with-ai-agents", "building-code-first-ai-agents-with-azure-ai-agent-service-cedric-vidal-microsoft", "building-interactive-uis-in-vs-code-with-mcp-apps-marlene-mhangami-liam-hampton-github", "building-protected-mcp-servers", "cooking-with-agents-in-vs-code", "creating-and-scaling-your-own-custom-copilots-with-azure-ai-studio", "don-t-let-the-llm-drive-ornella-bahidika-joel-allou-microsoft", "foundry-local-cutting-edge-ai-experiences-on-device-with-onnx-runtime-and-olive-emma-ning-micros", "from-writing-code-to-designing-systems-how-the-developer-role-is-changing-chris-noring-microsoft", "full-spec-mcp-hidden-capabilities-of-the-mcp-spec-harald-kirschner-microsoft-vs-code", "insights-from-snorkel-ai-running-azure-ai-infrastructure", "mind-the-gap-in-your-agent-observability", "multi-model-multimodal-and-multi-agent-innovations-in-azure-ai", "on-ai-and-knowledge", "pragmatic-ai-with-typechat", "rag-at-scale-production-ready-genai-apps-with-azure-ai-search", "real-world-development-with-github-copilot-and-vs-code-harald-kirschner-christopher-harrison", "running-ai-application-in-minutes-quick-start-with-ai-templates", "running-ai-application-in-minutes-quick-start-with-ai-templates-ee4087", "vibe-coding-at-scale-customizing-ai-assistants-for-enterprise-environments", "vibe-coding-at-scale-customizing-ai-assistants-for-enterprise-environments-harald-kirschner", "your-agent-failed-in-prod-good-luck-reproducing-it", "your-voice-agent-doesn-t-need-a-frontier-model" ] }, { "slug": "google-deepmind", "name": "Google DeepMind", "url": "https://ai.engineer/orgs/google-deepmind", "summary": null, "speakerSlugs": [ "logan-kilpatrick", "nicholas-kang", "michael-aaron", "philipp-schmid", "patrick-lober", "paige-bailey", "guillaume-vernade", "ian-ballantyne", "thor-schaeff", "sander-dieleman", "kat-kampf", "ammaar-reshi", "kevin-hou", "paige", "florina-muntenescu", "oli-gaymond", "cassidy-hardin", "omar-sanseviero", "mukund-sridhar", "aarush-selvan", "raia-hadsell", "kp-sawhney", "shrestha-basu-mallick", "benoit-schillings", "gus-martins", "brendan-o-donoghue", "jack-rae", "kathleen-kenealy" ], "talkSlugs": [ "a-year-of-gemini-progress-what-comes-next", "agentic-evaluations-at-scale-for-everybody", "ai-engineering-with-the-google-gemini-2-5-model-family", "any-to-any-building-native-multimodal-agents", "build-deploy-ai-powered-apps", "build-deploy-ai-powered-apps-18cc1b", "building-conversational-agents", "building-generative-image-video-models-at-scale", "building-in-the-gemini-era-kat-kampf-ammaar-reshi-google-deepmind", "defying-gravity", "don-t-ship-skills-without-evals", "from-transcription-to-live-music-gemini-s-audio-stack-thor-schaeff-google-deepmind", "frontier-feud", "gemini-nano-on-device-florina-muntenescu-oli-gaymond-google-deepmind", "gemma-4-deep-dive-cassidy-hardin-google-deepmind", "gemma-deepmind-s-family-of-open-models", "how-deep-research-works", "how-google-deepmind-is-researching-the-next-frontier-of-ai-for-gemini-raia-hadsell-vp-of-researc", "how-google-deepmind-runs-agents-at-scale-kp-sawhney-ian-ballantyne-google-deepmind", "let-s-go-bananas-with-genmedia", "milliseconds-to-magic-real-time-workflows-using-the-gemini-live-api-and-pipecat", "research-to-reality-with-google-deepmind", "sovereign-escape-velocity-ownership-with-open-models-gus-martins-and-ian-ballantyne-google-deepm", "text-diffusion-brendan-o-donoghue-google-deepmind", "thinking-deeper-in-gemini", "unveiling-the-latest-gemma-model-advancements", "veo-3-for-developers", "why-senior-engineers-struggle-to-build-ai-agents" ] }, { "slug": "anthropic", "name": "Anthropic", "url": "https://ai.engineer/orgs/anthropic", "summary": "Agent architecture and tool connectivity anchor Anthropic's recorded AIE appearances, with Barry Zhang on effective agents, Mahesh Murag on MCP, and Thariq Shihipar on the Claude Agent SDK. The archive also covers prompt experimentation, enterprise adoption, source-code security, and longer-running work, connecting implementation choices to the practical demands of deploying and supervising agents.", "speakerSlugs": [ "barry-zhang", "mahesh-murag", "alexander-bricken", "joe-bayley", "gagan-bhat", "isabella-kai-he", "jamie-neuwirth", "zack-witten", "thariq-shihipar", "boris-cherny", "cat-wu", "lance-martin", "katelyn-lesse", "kevin-bai", "mihir", "john", "david-soria-parra", "karan-sampath", "ash-prabaker", "andrew-wilson", "theodora-chu", "john-welsh", "eugene-yan", "alex-albert", "christopher-lovejoy", "karina-nguyen" ], "talkSlugs": [ "agent-skills", "anthropic-for-vps-of-ai", "anthropic-s-applied-ai-team-on-the-evolution-of-agentic-surfaces", "building-agents-with-mcp", "building-with-anthropic-s-claude-the-prompt-doctor-is-in", "claude-agent-sdk-workshop", "claude-code-the-evolution-of-agentic-coding", "claude-fable-claude-tag-and-anthropic-s-culture-cat-wu-thariq-shihipar-ft-simon-willison", "claude-for-long-horizon-tasks", "effective-ai-agents", "evolving-claude-apis-for-agents", "field-guide-to-fable", "forward-deployed-engineering-101", "frontier-feud", "future-of-mcp", "gateways-are-all-you-need", "how-to-build-agents-that-run-for-hours-without-losing-the-plot", "mcp-origins-and-requests-for-startups-theodora-chu-model-context-protocol-pm-anthropic", "remote-mcps-what-we-learned-from-shipping-john-welsh-anthropic", "using-llms-to-secure-source-code", "what-s-new-from-anthropic-and-what-s-next", "why-your-enterprise-tech-stack-isn-t-ready-for-ai-agents-and-what-to-build-instead", "writing-principles-for-task-tuned-prompt-engineering" ] }, { "slug": "openai", "name": "OpenAI", "url": "https://ai.engineer/orgs/openai", "summary": "Coding-agent reliability and production voice systems anchor OpenAI’s recorded AIE appearances, with Ryan Lopopolo on harness engineering and Anoop Kotha and Toki Sherbakov on voice architecture. Other sessions examine durable specifications, sandbox security, and multimodal interfaces, while Will Hang and Cathy Zhou explain training tool-using agents against custom rewards.", "speakerSlugs": [ "will-hang", "cathy-zhou", "anoop-kotha", "toki-sherbakov", "dominik-kundel", "vaibhav-srivastav", "katia-gil-guzman", "karina-nguyen", "greg-brockman", "abhishek-bhardwaj", "romain-huet", "jason-liu", "ilan-bigio", "bill-chen", "brian-fioca", "ryan-lopopolo", "prashant-mital", "fouad-matin", "logan-kilpatrick", "simon-fishman", "sean-grove", "peter-steinberger", "alexander-embiricos", "vinoth-govindarajan", "sean-dubois" ], "talkSlugs": [ "agent-reinforcement-fine-tuning", "building-effective-voice-agents", "building-voice-agents-with-openai", "codex-and-subagents", "codex-behind-the-harness", "creating-agents-that-co-create", "define-ai-engineer", "from-fork-to-fleet-designing-an-agent-sandbox-cloud-abhishek-bhardwaj-openai", "from-text-to-vision-to-voice-exploring-multimodality-with-openai", "full-workshop-setting-yourself-up-for-success-jason-liu-openai-codex", "function-calling-is-all-you-need", "future-proof-coding-agents", "harness-engineering", "model-maxxing-rft-dpo-sft-fine-tuning-with-openai-ilan-bigio-openai", "openai-for-vps-of-ai", "safety-and-security-for-code-executing-agents", "see-hear-speak-draw", "specifications-are-the-new-code", "state-of-the-claw", "the-golden-age-of-ai-engineering", "your-agent-didn-t-fail-your-harness-did", "your-realtime-ai-is-ngmi-sean-dubois-openai-kwindla-hultman-kramer-daily" ] }, { "slug": "neo4j", "name": "Neo4j", "url": "https://ai.engineer/orgs/neo4j", "summary": "GraphRAG, persistent agent memory, and enterprise context connect Neo4j’s recorded AIE appearances. Emil Eifrem introduces relationship-aware retrieval, Stephen Chin examines graph-grounded assistants, and Zach Blumenfeld demonstrates decision traces and lakehouse integration. Jesús Barrasa adds ontology-guided construction, while workshops and production discussions examine how these approaches work with real organizational data.", "speakerSlugs": [ "stephen-chin", "zach-blumenfeld", "ben-squire", "ryan-knight", "andreas-kollegger", "zaid-zaim", "emil-eifrem", "alex-gilmore", "michael-hunger", "jesus-barrasa", "philip-rathle" ], "talkSlugs": [ "agentic-graphrag-ai-s-logical-edge", "agentic-graphrag-simplifying-retrieval-across-structured-unstructured-data-zach-blumenfeld", "ai-on-your-lakehouse-context-comes-in-shapes-not-queries", "anchoring-enterprise-genai-with-knowledge-graphs", "connecting-the-dots-with-context-graphs", "context-engineering-connecting-the-dots-with-graphs-stephen-chin-neo4j", "context-graphs-for-explainable-decision-aware-ai-agents", "crabrag-why-automated-assistants-need-graph-memory-not-more-tokens", "graph-intelligence-enhance-reasoning-and-retrieval-using-graph-analytics", "graphrag", "intro-to-graphrag", "memory-masterclass-make-your-ai-agents-remember-what-they-do-mark-bain-aius", "practical-graphrag-making-llms-smarter-with-knowledge-graphs", "thinner-agents-on-a-smarter-substrate-the-ontology-based-semantic-layer", "understanding-ai-stakes-to-break-production-code", "why-your-agent-s-brain-needs-a-playbook-practical-wins-from-using-ontologies", "why-your-agents-need-decision-traces-not-just-documents-zach-blumenfeld-neo4j" ] }, { "slug": "aws", "name": "Amazon Web Services (AWS)", "url": "https://ai.engineer/orgs/aws", "summary": "At AIE, Suman Debnath demonstrated Strands agent development, Justin Muller examined evaluation failures, and Chintan Agrawal and Daniel Wirjo addressed conversational voice engineering. AWS-affiliated appearances also covered enterprise retrieval, browser automation, and coding assistants, offering practical entry points into how models connect to tools, data, and deployed applications.", "speakerSlugs": [ "justin-muller", "olivier-leplus", "antje-barth", "du-an-lightfoot", "banjo-obayomi", "derek-bingham", "mani-khanuja", "chintan-agrawal", "daniel-wirjo", "mike-chambers", "suman-debnath", "rohit-talluri", "elizabeth-fuentes-leone", "vikash-agrawal", "linda" ], "talkSlugs": [ "7-habits-of-highly-effective-generative-ai-evaluations", "ai-didn-t-kill-the-web-it-moved-in-olivier-leplus-aws-yohan-lasorsa-microsoft", "building-agents-at-cloud-scale-antje-barth-aws", "building-agents-with-amazon-nova-act-and-mcp-du-an-lightfoot-and-banjo-obayomi", "claude-plays-minecraft-introducing-a-real-world-serverless-ai-to-a-virtual-world", "data-is-your-differentiator-building-secure-and-tailored-ai-systems", "designing-voice-agents-for-real-conversations", "from-software-developer-to-ai-engineer", "introducing-strands-agents-an-open-source-ai-agents-sdk", "serving-voice-ai-at-scale-arjun-desai-cartesia-rohit-talluri-aws", "ship-it-building-production-ready-agents", "stop-ai-agent-hallucinations-5-techniques-production-patterns-elizabeth-fuentes-leone-aws", "supercharging-developer-workflow-with-amazon-q-developer", "using-spec-driven-development-for-production-workflows", "voicevision-rag-integrating-visual-document-intelligence-with-voice-response", "your-agent-is-wasting-tokens-and-you-don-t-know-it" ] }, { "slug": "github", "name": "GitHub", "url": "https://ai.engineer/orgs/github", "summary": "GitHub's recorded AIE appearances span Copilot product development with Mario Rodriguez, practical agent workflows with Jon Peck and Christopher Harrison, and team collaboration experiments with Maggie Appleton and Idan Gazit. The sessions connect everyday coding assistance to broader questions about tool access, security, developer control, and coordinating work across a team.", "speakerSlugs": [ "jon-peck", "christopher-harrison", "thomas-dohmke", "dave-burnison", "alex-malebranche", "dimitrios-philliou", "christina-warren", "harald", "rahul-pandita", "sarah-khalife", "maggie-appleton", "idan-gazit", "sam-morrow", "john-peck", "mario-rodriguez" ], "talkSlugs": [ "collaborating-with-agents-in-your-software-development-workflow-jon-peck-christopher-harrison-gi", "copilots-everywhere", "github-copilot-the-world-s-most-widely-adopted-ai-developer-tool", "github-copilot-the-world-s-most-widely-adopted-ai-developer-tool-repeat", "github-next-explorations", "github-s-ai-powered-security-platform", "one-developer-two-dozen-agents-zero-alignment-why-we-need-collaborative-ai-engineering", "piloting-agents-in-github-copilot", "real-world-development-with-github-copilot-and-vs-code-harald-kirschner-christopher-harrison", "real-world-mcps-in-github-copilot-agent-mode", "realtime-multiplayer-automation-and-you", "scaling-github-for-your-agents-sam-morrow-github", "the-agent-awakens-collaborative-development-with-copilot", "the-ai-evolution", "unlocking-ai-powered-devops-within-your-organization" ] }, { "slug": "arize", "name": "Arize", "url": "https://ai.engineer/orgs/arize", "summary": "Agent evaluation and failure diagnosis anchor Arize's recorded AIE appearances, with Aparna Dhinakaran and Dat Ngo examining how teams measure application behavior and investigate traces. Sessions featuring Sally-Ann DeLucia and Rachel Lee Nabors (RL Nabors) extend that engineering focus to long-session context management, on-device model selection, and browser interfaces for agents.", "speakerSlugs": [ "aparna-dhinakaran", "sally-ann-delucia", "fuad-ali", "dat-ngo", "aman-khan", "jason-lopatecki", "rachel-lee-nabors-rl-nabors", "laurie-voss" ], "talkSlugs": [ "break-it-til-you-make-it-building-the-self-improving-stack-for-ai-agents", "build-a-prompt-learning-loop", "engineering-better-evals-scalable-llm-evaluation-pipelines-that-work", "ensure-ai-agents-work-evaluation-frameworks-for-scaling-success", "from-signal-to-pr-anatomy-of-a-self-improving-agent", "frontier-results-on-device-rl-nabors-arize", "how-we-solved-context-management-in-agents-sally-ann-delucia", "llm-evals-that-work-irl", "llm-observability-evaluation-experimentation-platform-dat-ngo-arize-ai", "ship-real-agents-hands-on-evals-for-agentic-applications", "shipping-ai-that-works-an-evaluation-framework-for-pms", "the-future-of-evals-from-llm-as-a-judge-to-agent-as-a-judge-aparna-dhinakaran-arize-ai", "the-unreasonable-effectiveness-of-prompt-learning", "your-agent-is-an-infinite-canvas" ] }, { "slug": "nvidia", "name": "NVIDIA", "url": "https://ai.engineer/orgs/nvidia", "summary": "Inference performance and agent improvement anchor NVIDIA's recorded AIE appearances: Kyle Kranen examines serving architecture, Mozhgan Kabiri Chimeh demonstrates local benchmarking, and Sylendran Arunagiri explains production feedback loops. Lovina Dmello adds infrastructure security, giving readers a route through deployment choices, model specialization, and operational safeguards.", "speakerSlugs": [ "chris-alexiuk", "asma-beevi", "jensen-huang", "sylendran-arunagiri", "kyle-kranen", "mitesh-patel", "carter-abdallah", "mark-moyou", "mozhgan-kabiri-chimeh", "nader-khalil", "travis-bartley", "myungjong-kim", "byungjoong", "jaehan", "carter", "annika-brundyn", "aastha-jhunjhunwala", "ziv-ilan", "lovina-dmello" ], "talkSlugs": [ "compression-at-the-edge", "define-ai-engineer", "effective-ai-agents-need-data-flywheels-not-the-next-biggest-llm-sylendran-arunagiri-nvidia", "hacking-the-inference-pareto-frontier", "hybridrag-a-fusion-of-graph-and-vector-retrieval-to-enhance-data-interpretation", "local-models-trust-control-optimization", "mastering-llm-inference-optimization-from-theory-to-cost-effective-deployment", "running-llms-locally-practical-llm-performance-on-dgx-spark-mozhgan-kabiri-chimeh-nvidia", "state-of-the-union-why-local-why-now", "the-end-of-awkward-ai-transcriptions", "the-state-of-model-routing-nvidia-cognition-openrouter", "what-is-a-humanoid-foundation-model-an-introduction-to-gr00t-n1", "you-might-not-need-50-diffusion-steps", "your-llm-stack-is-a-2008-database-with-better-marketing" ] }, { "slug": "braintrust", "name": "Braintrust", "url": "https://ai.engineer/orgs/braintrust", "summary": "AI evaluation and agent observability anchor Braintrust’s recorded AIE appearances: Doug Guthrie introduces evaluation workflows, Ankur Goyal examines dataset and scorer design, and Phil Hetzel connects expert judgment with production traces. Ameya Bhatawdekar extends the discussion to evolving agent architectures, while joint workshops examine how evaluation fits into application development.", "speakerSlugs": [ "phil-hetzel", "doug-guthrie", "carlos-esteban", "doug", "ankur-goyal", "giran-moodley", "manu-goyal", "ameya-bhatawdekar" ], "talkSlugs": [ "does-genai-belong-to-data-scientists", "evals-101-doug-guthrie-braintrust", "evals-workshop-mastering-ai-evaluation-from-playground-to-production", "five-hard-earned-lessons-about-evals", "how-agent-o11y-differs-from-traditional-o11y", "how-to-build-world-class-ai-products-sarah-sachs-notion-and-carlos-esteban-braintrust", "how-zapier-builds-ai-products-and-features-with-the-help-of-braintrust", "shipping-complex-ai-applications-braintrust-trainline", "the-future-of-evals-ankur-goyal-braintrust", "the-maturity-phases-of-running-evals", "why-building-eval-platforms-is-hard", "why-should-anyone-care-about-evals", "your-agent-evolved-your-evals-didn-t" ] }, { "slug": "google", "name": "Google", "url": "https://ai.engineer/orgs/google", "summary": "On-device agents, browser interfaces, and AI education connect Google’s recorded AIE appearances. Cormac Brick and Chintan Parikh discuss local deployment, Michael Hablich and Tara Agyemang examine tools for browser agents, and Stefania Druga addresses learner agency. Rukma Sen adds an introduction to agent architecture and enterprise workflows through Vertex AI.", "speakerSlugs": [ "chintan-parikh", "weiyi-wang", "michael-hablich", "cormac-brick", "petra", "rukma-sen", "tara-agyemang", "stefania-druga" ], "talkSlugs": [ "accelerating-ai-on-edge-chintan-parikh-and-weiyi-wang-google-deepmind", "building-agent-interfaces-lessons-from-chrome-devtools-mcp-for-agents", "from-46-to-90-fine-tuning-tiny-llms-for-on-device-agents-cormac-brick-google", "frontier-feud", "level-up-your-automation-game-agentic-workflows-on-vertex-ai", "the-agent-ready-web-simplify-user-actions-with-webmcp", "the-multimodal-future-of-education", "tlms-tiny-llms-and-agents-on-edge-devices-with-litert-lm", "tools-for-the-next-generation-of-ai-engineers", "why-large-tiny-lms-agents-on-edge-robotics" ] }, { "slug": "workos", "name": "WorkOS", "url": "https://ai.engineer/orgs/workos", "summary": "Agent identity and dependable execution connect WorkOS appearances at AI Engineer. Michael Grinich, Garrett Galow, and Tobin South discuss authorization and enterprise access, while Nick Nisi and Zack Proser examine coding workflows, reusable skills, and human attention. Galow also presents an internal workspace for answering business questions through executable queries and reusable widgets.", "speakerSlugs": [ "nick-nisi", "zack-proser", "michael-grinich", "garrett-galow", "tobin-south" ], "talkSlugs": [ "agents-redacted-location-address-access-redacted-location-address-and-the-future-of-machine-iden", "ai-pipelines-and-agents-in-pure-typescript-with-mastra-ai", "ciam-for-ai-authn-authz-for-agents-michael-grinich-ceo-of-workos", "one-login-to-rule-them-all-cross-app-access-for-mcp", "pruning-agent-skills", "skills-at-scale", "what-does-enterprise-ready-mcp-mean", "why-can-t-anyone-answer-questions-about-the-business", "your-attention-is-the-bottleneck-not-your-agents-zack-proser-workos" ] }, { "slug": "cloudflare", "name": "Cloudflare", "url": "https://ai.engineer/orgs/cloudflare", "summary": "Cloudflare's recorded AIE appearances examine how agents retain state, access tools, and execute generated code. Rita Kozlov introduces agent building blocks, while Sunil Pai and Matt Carey examine code execution and API discovery. Kenton Varda and Harshil Agrawal address isolation for personal applications and untrusted generated programs.", "speakerSlugs": [ "lizzie-siegle", "rita-kozlov", "sunil-pai", "kenton-varda", "matt-carey", "harshil-agrawal" ], "talkSlugs": [ "agents-redacted-location-address-access-redacted-location-address-and-the-future-of-machine-iden", "building-agents-the-hard-parts", "code-mode-let-the-code-do-the-talking", "gadgets-personal-app-vibe-coding-that-is-actually-safe", "mcp-mega-context-problem-matt-carey", "why-and-how-you-need-to-sandbox-ai-generated-code-harshil-agrawal-cloudflare", "why-eval-is-the-next-great-compute-primitive-sunil-pai-matt-carey-cloudflare" ] }, { "slug": "daily", "name": "Daily", "url": "https://ai.engineer/orgs/daily", "summary": "Realtime voice systems anchor Daily's AIE appearances: Kwindla Hultman Kramer examines responsive agents and deployment, while Mark Backman and Aleix teach Pipecat development. Chad Bailey joins a discussion of conversational video with Tavus. Across these sessions, practical audio pipelines connect to broader questions about how voice, tools, and context shape software interfaces.", "speakerSlugs": [ "mark-backman", "aleix", "kwindla-hultman-kramer", "kwindla-kramer", "chad-bailey" ], "talkSlugs": [ "full-workshop-realtime-voice-ai-mark-backman-daily", "how-to-build-the-world-s-fastest-voice-bot", "milliseconds-to-magic-real-time-workflows-using-the-gemini-live-api-and-pipecat", "pipecat-cloud-enterprise-voice-agents-built-on-open-source", "realtime-conversational-video-with-pipecat-and-tavus-chad-bailey-and-brian-johnson-daily-tavus", "the-new-primitives-building-ai-native-software", "your-realtime-ai-is-ngmi-sean-dubois-openai-kwindla-hultman-kramer-daily" ] }, { "slug": "hugging-face", "name": "Hugging Face", "url": "https://ai.engineer/orgs/hugging-face", "summary": "At AIE, Ben Burtenshaw examines coding agents for AI systems engineering, Arek Borucki explains Hub infrastructure, and Andres Marafioti demonstrates conversational robotics. Hugging Face’s recorded appearances also include Niels Rogge on research outreach automation, Merve Noyan on agent tooling and compression, and Thom Wolf in a joint cybersecurity benchmark discussion.", "speakerSlugs": [ "merve-noyan", "niels-rogge", "andres-marafioti", "arek-borucki", "thom-wolf", "ben-burtenshaw" ], "talkSlugs": [ "compression-at-the-edge", "how-i-automate-my-own-job-at-hugging-face-using-agents", "reachy-mini-the-300-open-source-robot-you-can-actually-hack-andres-marafioti-hugging-face", "self-training-agents-hermes-agent-hf-traces-skills-mcp-finetuning-merve-noyan-hugging-face", "serving-2-million-models-without-melting-scaling-the-hugging-face-hub", "training-frontier-models-to-out-think-hackers-uri-rolls-arithmetic-thom-wolf-hugging-face", "your-coding-agent-should-do-ai-system-engineering" ] }, { "slug": "amazon", "name": "Amazon", "url": "https://ai.engineer/orgs/amazon", "summary": "Amazon-affiliated AIE appearances connect production LLM engineering with search, software development, and interface design. Eugene Yan covers evaluation and recommendation systems, Al Harris demonstrates Kiro's requirements-to-implementation workflow, and Bala Ramdoss examines how agent output becomes usable UI. Joint discussions add perspectives on Copilot adoption and the operational work beyond demos.", "speakerSlugs": [ "bala-ramdoss", "eugene-yan", "al-harris" ], "talkSlugs": [ "agent-output-is-not-ux-rendering-layer-your-llm-pipeline-is-missing-bala-ramdoss-amazon-lens", "building-blocks-for-llm-systems-products", "copilots-everywhere", "recsys-keynote-improving-recommendation-systems-search-in-the-age-of-llms", "spec-driven-development-agentic-coding-at-faang-scale-and-quality-al-harris-amazon-kiro", "what-we-learned-from-a-year-of-building-with-llms" ] }, { "slug": "cursor", "name": "Cursor", "url": "https://ai.engineer/orgs/cursor", "summary": "Cursor's recorded AIE appearances connect coding-model development with the workflows and organizations that use agents. Lee Robinson discusses Composer training and recursive improvement; Naman Jain examines evaluation across codebases. Eric Zakariasson and David Gomes address autonomous workflows and implementation tradeoffs, while Pauline Brunet explains how to scope enterprise engineering engagements.", "speakerSlugs": [ "lee-robinson", "eric-zakariasson", "naman-jain", "pauline-brunet", "david-gomes" ], "talkSlugs": [ "building-cursor-composer", "building-your-own-software-factory", "coding-evals-from-code-snippets-to-codebases-naman-jain-cursor", "forward-deployed-engineering-at-cursor-pauline-brunet", "recursive-model-improvement", "replacing-12k-loc-with-a-200-loc-skill" ] }, { "slug": "latent-space", "name": "Latent.Space", "url": "https://ai.engineer/orgs/latent-space", "summary": "AI engineering roles, reasoning models, and agent design connect Latent.Space’s recorded AIE appearances. swyx examines the discipline’s boundaries and practical direction, Vibhu Sapra explains DeepSeek-V3 and DeepSeek-R1, and Alessio Fanelli joins Clay Bavor for a discussion of the AI architect’s responsibility for customer-facing agents. These sessions span technical foundations, application design, and organizational practice.", "speakerSlugs": [ "swyx", "vibhu-sapra", "alessio-fanelli" ], "talkSlugs": [ "ai-engineering-without-borders", "designing-ai-intensive-applications", "latent-space-paper-club-aiewf-special-edition-test-of-time-deepseek-r1-v3-vibhu-sapra", "rise-of-the-ai-architect-clay-bavor-and-alessio-fanelli", "the-1-000x-ai-engineer-swyx", "why-agent-engineering" ] }, { "slug": "mongodb", "name": "MongoDB", "url": "https://ai.engineer/orgs/mongodb", "summary": "At AIE, Ben Flast and Frank Liu cover vector retrieval and search quality, Richmond Alake examines persistent agent memory, and Apoorva Joshi addresses multimodal agents and production design. MongoDB’s recorded appearances connect database mechanics with application architecture, including a research-agent workshop credited to Apoorva Joshi and Ben Perlmutter.", "speakerSlugs": [ "apoorva-joshi", "richmond-alake", "ben-flast", "ben-perlmutter", "frank-liu" ], "talkSlugs": [ "ai-system-design-from-idea-to-production", "architecting-agent-memory-principles-patterns-and-best-practices", "building-multimodal-ai-agents-from-scratch", "rag-and-the-mongodb-document-model", "the-a-to-z-of-building-ai-agents", "the-state-of-ai-powered-search-and-retrieval" ] }, { "slug": "vercel", "name": "Vercel", "url": "https://ai.engineer/orgs/vercel", "summary": "AI SDK workflows and dependable agent execution connect Vercel’s recorded AIE appearances by Nico Albanese and Peter Wielander. Ido Pesok addresses evaluation, while Hassan El Mghari describes weekend application building and distribution. Malte Ubl broadens the discussion to software economics and infrastructure for agents, linking practical implementation with changes in who uses applications.", "speakerSlugs": [ "nico-albanese", "peter-wielander", "ido-pesok", "malte-ubl", "hassan-el-mghari" ], "talkSlugs": [ "ai-sdk-v6", "building-durable-agents-with-workflow-devkit-ai-sdk", "evals-are-not-unit-tests", "the-new-application-layer", "the-weekend-ai-engineer", "vercel-ai-sdk-masterclass-from-fundamentals-to-deep-research" ] }, { "slug": "ai-engineer", "name": "AI Engineer", "url": "https://ai.engineer/orgs/ai-engineer", "summary": "Agent workflows and the quality of AI-assisted work anchor Shawn \"swyx\" Wang's recorded AIE appearances, including conference operations and a preview of World's Fair programming. Shared sessions with Gergely Orosz and Peter Steinberger extend the coverage to software engineering incentives and the operational challenges of open agents.", "speakerSlugs": [ "swyx" ], "talkSlugs": [ "6-things-to-know-about-aie-world-s-fair-2026", "agents-for-everything-else-swyx", "no-more-slop-swyx", "software-engineering-ai", "state-of-the-claw" ] }, { "slug": "augment-code", "name": "Augment Code", "url": "https://ai.engineer/orgs/augment-code", "summary": "Coding agents that extend their own codebase, investigate legacy software, and take delegated engineering tasks anchor Augment Code's recorded AIE appearances. Colin Flaherty covers agent construction, Eric Hou demonstrates supervised delegation, Matt Ball joins a legacy-code demonstration, and Chris Kelly examines the responsibilities that remain when generating code becomes easier.", "speakerSlugs": [ "colin-flaherty", "colin", "eric-hou", "matt-ball", "chris-kelly" ], "talkSlugs": [ "building-self-coding-agents", "frontier-feud", "mentoring-the-machine", "to-the-moon-navigating-deep-context-in-legacy-code-with-augment-agent", "vibes-won-t-cut-it" ] }, { "slug": "elevenlabs", "name": "ElevenLabs", "url": "https://ai.engineer/orgs/elevenlabs", "summary": "Voice agents anchor ElevenLabs’ recorded AIE appearances: Thor Schaeff teaches multilingual conversation design, Luke Harries demonstrates adding speech to existing agents, and Joe Reeve turns statue photographs into character conversations. Boris Starkov connects an undocumented phone to a conversational agent, while Angelos Perivolaropoulos teaches language-model training from scratch on local hardware.", "speakerSlugs": [ "thor-schaeff", "paul", "luke-harries", "joe-reeve", "boris-starkov", "angelos-perivolaropoulos" ], "talkSlugs": [ "building-conversational-ai-agents-thor-schaeff-elevenlabs", "give-your-chat-agent-a-voice", "how-to-talk-to-statues-joe-reeve-elevenlabs", "reverse-engineering-a-viking-voip-phone-protocol-with-claude-code-boris-starkov-elevenlabs", "training-an-llm-from-scratch-locally" ] }, { "slug": "factory", "name": "Factory", "url": "https://ai.engineer/orgs/factory", "summary": "Autonomous software engineering connects Factory's AIE appearances: Eno Reyes discusses Droids, mechanically verifiable codebases, organizational context, and enterprise deployment. Luke Alvoeiro presents Missions, separating orchestration, implementation, and validation. Together, these talks examine the engineering conditions needed for agents to carry work beyond code generation into sustained software delivery.", "speakerSlugs": [ "eno-reyes", "luke-alvoeiro" ], "talkSlugs": [ "building-reliable-agentic-systems", "how-forward-deployed-engineering-is-done-at-factory", "making-codebases-agent-ready", "production-multi-agent-architecture", "ship-production-software-in-minutes-not-months" ] }, { "slug": "humanlayer", "name": "HumanLayer", "url": "https://ai.engineer/orgs/humanlayer", "summary": "At AI Engineer, Dex Horthy and Kyle Mistele examine how to make agents useful in complex codebases while preserving engineering judgment. Their recorded talks connect explicit application control, context engineering, and measurable coding workflows, with attention to the review burden and architectural risks that can accompany faster implementation.", "speakerSlugs": [ "dex-horthy", "kyle-mistele" ], "talkSlugs": [ "12-factor-agents", "context-engineering-for-complex-codebases", "harness-engineering-is-not-enough-why-software-factories-fail", "loop-engineering-from-first-principles", "the-great-loops-debate-dex-horthy-geoff-huntley-ian-livingstone-greg-pstrucha-moderated-by-allie" ] }, { "slug": "llamaindex", "name": "LlamaIndex", "url": "https://ai.engineer/orgs/llamaindex", "summary": "Jerry Liu and Laurie Voss's recorded LlamaIndex appearances at AI Engineer address how retrieval systems develop into knowledge assistants and document agents. Liu covers evaluation, enterprise architecture, and document automation; Voss examines when agents justify their complexity and how production workflow patterns apply to unstructured data and software integrations.", "speakerSlugs": [ "jerry-liu", "laurie-voss" ], "talkSlugs": [ "building-ai-agents-that-actually-automate-knowledge-work", "effective-agent-design-patterns-in-production", "future-of-knowledge-assistants", "production-ready-rag", "rag-for-vps-of-ai" ] }, { "slug": "modal", "name": "Modal", "url": "https://ai.engineer/orgs/modal", "summary": "At AIE, Charles Frye connects GPU hardware to practical self-hosted inference, while Nan Jiang examines reinforcement learning across datacenters. Benjamin Cowen discusses when specialized applications benefit from custom models and the infrastructure they require. A joint session featuring Frye broadens the archive to evaluations, monitoring, and the work of moving beyond demos.", "speakerSlugs": [ "charles-frye", "nan-jiang", "benjamin-cowen" ], "talkSlugs": [ "how-fast-are-llm-inference-engines-anyway", "taking-reinforcement-learning-cross-datacenter", "what-every-ai-engineer-needs-to-know-about-gpus", "what-lies-beneath-the-api-benjamin-cowen-modal", "what-we-learned-from-a-year-of-building-with-llms" ] }, { "slug": "prime-intellect", "name": "Prime Intellect", "url": "https://ai.engineer/orgs/prime-intellect", "summary": "At AI Engineer, Will Brown's talks connect agent training to reinforcement learning, reusable environments, and post-training infrastructure, including tasks whose rewards are difficult to verify. Vincent Weisser participates in a joint discussion of local and open models, extending the archive's focus to ownership, privacy, model provenance, and practical access to inference.", "speakerSlugs": [ "vincent-weisser", "will-brown" ], "talkSlugs": [ "local-models-trust-control-optimization", "modern-post-training-a-deep-dive-will-brown-prime-intellect", "reinforcement-learning-without-verifiable-rewards-will-brown-prime-intellect", "rl-environments-at-scale", "training-agentic-reasoners" ] }, { "slug": "qodo", "name": "Qodo", "url": "https://ai.engineer/orgs/qodo", "summary": null, "speakerSlugs": [ "dedy-kredo", "itamar-friedman", "nupur-sharma" ], "talkSlugs": [ "move-fast-break-nothing", "the-last-human-code-review-building-trust-in-ai-generated-code", "the-state-of-ai-code-quality-hype-vs-reality", "vibe-coding-with-confidence", "why-more-context-makes-your-agent-dumber-and-what-to-do-about-it" ] }, { "slug": "temporal", "name": "Temporal", "url": "https://ai.engineer/orgs/temporal", "summary": "Durable agent execution anchors Temporal's recorded AIE appearances: Cornelia Davis demonstrates workflow integration and asynchronous tools, while Mason Egger and Preeti Somal examine distributed-system reliability. Johann Schleier-Smith extends the discussion to verification of LLM-generated software, connecting agent infrastructure with broader engineering methods for preventing and recovering from failures.", "speakerSlugs": [ "cornelia-davis", "mason-egger", "preeti-somal", "johann-schleier-smith" ], "talkSlugs": [ "building-durable-production-ready-agents-with-openai-sdk-and-temporal", "events-are-the-wrong-abstraction-for-your-ai-agents", "mcp-tasks-async-why-the-heck-aren-t-any-agents-supporting-mcp-tasks-async", "scaling-ai-agents-without-breaking-reliability", "vision-zero-bugs" ] }, { "slug": "together-ai", "name": "Together AI", "url": "https://ai.engineer/orgs/together-ai", "summary": "Together AI's AIE appearances connect infrastructure constraints with application development: Max Ryabinin examines long-context training, Rishabh Bhargava addresses voice latency, and Hassan El Mghari demonstrates app building and interface refinement. A joint discussion featuring Dan Fu and MiniMax's Olive Song adds multimodal model serving, agent benchmarks, and GPU inference challenges.", "speakerSlugs": [ "dan-fu", "rishabh-bhargava", "max-ryabinin", "hassan-el-mghari" ], "talkSlugs": [ "agents-at-scale-inside-minimax-s-model-and-the-infrastructure-behind-it", "engineering-voice-agents-latency-quality-and-scale", "road-to-5-million-tokens-breaking-barriers-in-long-context-training-max-ryabinin-together-ai", "the-missing-layer-design-taste-in-ai-agents", "using-oss-models-to-build-ai-apps-with-millions-of-users" ] }, { "slug": "unsloth", "name": "Unsloth", "url": "https://ai.engineer/orgs/unsloth", "summary": "Daniel Han’s Unsloth appearances at AI Engineer connect open-model debugging with reinforcement learning and efficient deployment. His talks and workshops examine how implementation details, reward functions, and training optimizations affect model behavior; a shared panel discussion extends the coverage to compression for local and edge use.", "speakerSlugs": [ "daniel-han" ], "talkSlugs": [ "compression-at-the-edge", "fixing-bugs-in-gemma-llama-phi-3", "low-level-technicals-of-llms", "rl-quantization-agents", "special-topics-in-kernels-rl-reward-hacking-in-agents" ] }, { "slug": "windsurf", "name": "Windsurf", "url": "https://ai.engineer/orgs/windsurf", "summary": "Windsurf’s recorded AIE appearances feature Kevin Hou, Eashan Sinha, and Sam Fertig on repository retrieval, developer intent, and collaborative coding agents. Their talks trace the move from autocomplete to shared human-agent workflows, connecting context selection with practical questions about latency, terminal approvals, tool use, and evaluation of engineering tasks.", "speakerSlugs": [ "kevin-hou", "eashan-sinha", "sam-fertig" ], "talkSlugs": [ "agents-are-built-at-the-fringe-getting-from-90-to-100", "embeddings-are-stunting-agents-how-codeium-breaks-through-the-ceiling-for-retrieval", "mastering-engineering-flow-with-windsurf", "the-eyes-are-the-context-window-to-the-soul-how-windsurf-gets-to-know-you", "windsurf-everywhere-doing-everything-all-at-once" ] }, { "slug": "amazon-agi-lab", "name": "Amazon AGI Lab", "url": "https://ai.engineer/orgs/amazon-agi-lab", "summary": null, "speakerSlugs": [ "aditya-khandelwal", "gaurav-mishra", "antje-barth", "danielle-perszyk" ], "talkSlugs": [ "amazon-agi", "from-rl-to-irl-gaurav-mishra-amazon-agi-lab", "perception-agents-antje-barth-amazon-agi-lab", "useful-general-intelligence" ] }, { "slug": "anterior", "name": "Anterior", "url": "https://ai.engineer/orgs/anterior", "summary": "Healthcare AI evaluation and deployment connect Anterior's recorded AIE appearances. Christopher Lovejoy examines clinician feedback and medical decision review; Anuj Iravane addresses synthetic records when retention restrictions constrain evaluation data. Saul Howard joins a discussion of enterprise infrastructure, extending the focus from model quality to the controls needed for healthcare agents in production.", "speakerSlugs": [ "anuj-iravane", "christopher-lovejoy", "saul-howard" ], "talkSlugs": [ "don-t-be-data-poor", "make-your-llm-app-a-domain-expert-how-to-build-an-llm-native-expert-system", "mission-critical-evals-at-scale-learnings-from-100-000-medical-decisions", "why-your-enterprise-tech-stack-isn-t-ready-for-ai-agents-and-what-to-build-instead" ] }, { "slug": "baseten", "name": "Baseten", "url": "https://ai.engineer/orgs/baseten", "summary": "Production inference connects Baseten's recorded AIE appearances: Philip Kiely and Pankaj Gupta cover TensorRT-LLM deployment, while Kiely and Yineng Zhang introduce SGLang serving. Kiely's voice-model talk examines real-time audio performance, and Amir Haghighat discusses the infrastructure and operational expertise involved in building enterprise systems with open models.", "speakerSlugs": [ "philip-kiely", "pankaj-gupta", "yineng-zhang", "amir-haghighat" ], "talkSlugs": [ "from-model-weights-to-api-endpoint-with-tensorrt-llm", "introduction-to-llm-serving-with-sglang", "optimizing-inference-for-voice-models-in-production", "the-rise-of-open-models-in-the-enterprise" ] }, { "slug": "cline", "name": "Cline", "url": "https://ai.engineer/orgs/cline", "summary": "Coding-agent design, evaluation, and developer choice connect Cline's recorded AIE appearances. Ara Khan examines agent maturity and benchmark reliability; Nik Pash discusses frontier models and reinforcement-learning environments; Saoud Rizwan considers open-source collaboration under pressure from AI-generated contributions. Together, these talks connect practical engineering decisions with the costs and constraints of relying on model providers.", "speakerSlugs": [ "ara-khan", "nik-pash", "saoud-rizwan" ], "talkSlugs": [ "don-t-build-slop-4-levels-of-ai-agent-maturity", "evals-are-broken-use-them-anyway", "hard-won-lessons-from-building-effective-ai-coding-agents", "open-source-is-dead-long-live-open-source" ] }, { "slug": "cognition", "name": "Cognition", "url": "https://ai.engineer/orgs/cognition", "summary": "Devin’s software engineering workflows anchor Cognition’s recorded AIE appearances: Scott Wu demonstrates application building and explains the move toward broader autonomous tasks, while Jia Wu discusses enterprise deployment and customer outcomes. Walden participates in a joint panel examining model routing, extending the reading path to the infrastructure choices behind agent applications.", "speakerSlugs": [ "scott-wu", "jia-wu", "walden" ], "talkSlugs": [ "devin-2-0-and-the-future-of-swe", "how-forward-deployed-engineering-is-done-at-cognition", "the-making-of-devin", "the-state-of-model-routing-nvidia-cognition-openrouter" ] }, { "slug": "krea", "name": "Krea", "url": "https://ai.engineer/orgs/krea", "summary": "Krea's AIE appearances connect human judgments of generated images with the data and infrastructure used to build them. Diego Rodriguez examines aesthetic evaluation, Sangwu Lee describes Krea 2's training choices, and Gabriel Jorge Menezes explains distributed training and serving. A shared startup showcase also includes Krea's creative suite.", "speakerSlugs": [ "gabriel-jorge-menezes", "diego-rodriguez", "sangwu-lee" ], "talkSlugs": [ "infra-behind-krea-2-how-to-train-and-serve-at-scale", "perceptual-evaluations-evals-for-aesthetics-diego-rodriguez-krea-ai", "the-next-unicorns-7-top-ai-startups-from-the-hf0-residency", "training-krea-2-what-matters-in-generative-model-training" ] }, { "slug": "langchain", "name": "LangChain", "url": "https://ai.engineer/orgs/langchain", "summary": "At AIE, Harrison Chase, Lance Martin, and Vivek Trivedy discuss building agents with context, explicit control flow, and observable execution. Their recorded LangChain appearances connect application architecture and hands-on testing with enterprise adoption and learning from production failures, giving readers several entry points into the engineering work around autonomous systems.", "speakerSlugs": [ "harrison-chase", "lance-martin", "vivek-trivedy" ], "talkSlugs": [ "3-ingredients-for-building-reliable-enterprise-agents", "architecting-and-testing-controllable-agents", "building-context-aware-reasoning-applications-with-langchain-and-langsmith", "improving-agents-is-a-data-mining-problem" ] }, { "slug": "notion", "name": "Notion", "url": "https://ai.engineer/orgs/notion", "summary": "Notion's recorded AIE appearances connect AI product reliability with experimental interfaces and human understanding. Sarah Sachs discusses evaluation practices and model economics; Linus Lee demonstrates ways to manipulate and reconstruct content through embeddings; Geoffrey Litt examines how developers can understand software produced with agents as code generation becomes easier.", "speakerSlugs": [ "sarah-sachs", "linus-lee", "geoffrey-litt" ], "talkSlugs": [ "how-to-build-world-class-ai-products-sarah-sachs-notion-and-carlos-esteban-braintrust", "notion-s-token-town", "the-hidden-life-of-embeddings", "understanding-is-the-new-bottleneck" ] }, { "slug": "pydantic", "name": "Pydantic", "url": "https://ai.engineer/orgs/pydantic", "summary": "Samuel Colvin's Pydantic appearances at AI Engineer examine how to build, connect, and improve AI agents. His demonstrations move from typed extraction and validation to crash recovery with Temporal, communication through MCP, and production optimization with evaluation datasets. Logfire provides visibility into agent execution, model calls, timing, and costs across these sessions.", "speakerSlugs": [ "samuel-colvin" ], "talkSlugs": [ "from-stateless-nightmares-to-durable-agents", "human-seeded-evals-samuel-colvin-pydantic", "mcp-is-all-you-need", "playground-in-prod-optimising-agents-in-production-environments" ] }, { "slug": "sentry", "name": "Sentry", "url": "https://ai.engineer/orgs/sentry", "summary": "Code comprehension and agent correctness connect Sentry’s recorded AIE appearances: Priscila Andre de Oliveira presents repository understanding, Daniel Szoke examines Rust’s feedback for coding agents, and David Cramer assesses MCP integration tradeoffs. Greg Pstrucha participates in a debate about whether autonomous coding loops deliver reliable engineering outcomes.", "speakerSlugs": [ "priscila-andre-de-oliveira", "david-cramer", "greg-pstrucha", "daniel-szoke" ], "talkSlugs": [ "comprehend-first-code-later-the-ai-skill-i-rely-on-daily", "mcp-is-not-good-yet-david-cramer-sentry", "the-great-loops-debate-dex-horthy-geoff-huntley-ian-livingstone-greg-pstrucha-moderated-by-allie", "why-rust-is-the-ideal-language-for-vibe-coding" ] }, { "slug": "snorkel-ai", "name": "Snorkel AI", "url": "https://ai.engineer/orgs/snorkel-ai", "summary": "Agent evaluation and training connect Snorkel AI's recorded AIE appearances by Rustem Feyzkhanov, Kobie Crawford, and Vincent Chen. Their talks cover simulations built from production traces, task fidelity in reinforcement learning, financial-analysis tool use, and benchmark design, linking the quality of evaluation environments to the behaviors that training rewards and testing reveals.", "speakerSlugs": [ "rustem-feyzkhanov", "kobie-crawford", "vincent-chen" ], "talkSlugs": [ "from-agent-traces-to-agent-simulations-rustem-feyzkhanov-snorkel-ai", "stop-making-models-bigger-make-them-behave-kobie-crawford-snorkel", "task-fidelity-scaling-laws-kobie-crawford-snorkel-ai", "the-art-science-of-benchmarking-agents" ] }, { "slug": "sourcegraph", "name": "Sourcegraph", "url": "https://ai.engineer/orgs/sourcegraph", "summary": "Coding AI capability, developer adoption, and measurable business impact connect Sourcegraph’s recorded AIE appearances. Ado Kukic presents a maturity model for coding automation, Quinn Slack examines recurring tool use, and Beyang Liu discusses evaluation frameworks and joins Bruno Passos for an enterprise software-development case study.", "speakerSlugs": [ "beyang-liu", "quinn-slack", "ado-kukic" ], "talkSlugs": [ "building-ai-agents-with-real-roi-in-the-enterprise-sdlc", "the-ai-emperor-has-no-daus-why-most-devs-still-don-t-use-code-ai", "the-code-ai-maturity-model-and-what-it-means-for-you", "the-roi-of-ai-why-you-need-eval-frameworks" ] }, { "slug": "uc-berkeley", "name": "UC Berkeley", "url": "https://ai.engineer/orgs/uc-berkeley", "summary": null, "speakerSlugs": [ "frank-coyle", "parth-asawa", "shreya-shankar" ], "talkSlugs": [ "anthropic-s-cca-exam-as-a-field-guide-for-agentic-engineering", "beyond-static-intelligence-evaluating-continual-learning", "ontologies-for-agentic-systems", "what-we-learned-from-a-year-of-building-with-llms" ] }, { "slug": "weights-biases", "name": "Weights & Biases", "url": "https://ai.engineer/orgs/weights-biases", "summary": null, "speakerSlugs": [ "alex-volkov", "lukas-biewald" ], "talkSlugs": [ "judging-llms", "productionizing-genai-models-lessons-from-the-world-s-best-ai-teams", "the-state-of-mcp-observability-observable-tools-alex-volkov-and-benjamin-eckel-weights-biases-an", "the-z-l-continuum-should-ai-engineers-still-read-code" ] }, { "slug": "ai-hero", "name": "AI Hero", "url": "https://ai.engineer/orgs/ai-hero", "summary": "Matt Pocock's AI Hero appearances at AI Engineer examine how to make coding agents useful within disciplined software development. His talks connect requirements interviews and autonomous implementation with software design fundamentals and agent-skill construction, keeping human judgment central to architecture, product decisions, and verification of generated code.", "speakerSlugs": [ "matt-pocock" ], "talkSlugs": [ "ai-coding-workflow", "building-great-agent-skills", "software-fundamentals-for-ai-coding" ] }, { "slug": "all-hands-ai", "name": "All Hands AI", "url": "https://ai.engineer/orgs/all-hands-ai", "summary": null, "speakerSlugs": [ "robert-brennan", "ray-myers" ], "talkSlugs": [ "automating-large-scale-refactors-with-parallel-agents", "software-development-agents-what-works-and-what-doesn-t", "the-many-ends-of-programming" ] }, { "slug": "amplify-partners", "name": "Amplify Partners", "url": "https://ai.engineer/orgs/amplify-partners", "summary": "Barr Yaron’s AIE appearances connect practitioner survey findings with questions about deploying and operating AI systems. The Amplify Partners archive covers model adoption, prompt management, multimodal use, and agent controls across the 2025 and 2026 reports, alongside Frontier Feud, a hosted contest featuring participants’ predictions about AI development.", "speakerSlugs": [ "barr-yaron" ], "talkSlugs": [ "frontier-feud", "the-2025-ai-engineering-report-barr-yaron-amplify-partners", "the-2026-state-of-ai-engineering-barr-yaron-amplify-partners" ] }, { "slug": "applied-compute", "name": "Applied Compute", "url": "https://ai.engineer/orgs/applied-compute", "summary": "Enterprise reinforcement learning connects Applied Compute’s recorded AIE appearances: Rhythm Garg and Linden Li examine training efficiency, Raymond Feng discusses learning from extended enterprise interactions, and Samuel Denton compares continual-learning approaches. Together, these talks connect the systems needed to run training with the environments and feedback needed to improve agents.", "speakerSlugs": [ "samuel-denton", "rhythm-garg", "linden-li", "raymond-feng" ], "talkSlugs": [ "bringing-continual-learning-into-enterprises", "efficient-reinforcement-learning", "learning-on-the-job-the-future-of-post-training" ] }, { "slug": "bright-data", "name": "Bright Data", "url": "https://ai.engineer/orgs/bright-data", "summary": "Reliable web retrieval and reusable context pipelines connect Bright Data’s recorded AIE appearances. Rafael Levi examines misleading search results and demonstrates agent-generated scrapers, while Omer Primor discusses Context-as-a-Service and retrieval economics. Together, these talks address how agents obtain fresh web data, maintain collection workflows, and weigh purchased context against owned infrastructure.", "speakerSlugs": [ "rafael-levi", "omer-primor" ], "talkSlugs": [ "from-mcp-to-scale-pipelines-that-build-themselves", "the-rise-of-caas-context-as-a-service-for-agentic-ai", "your-agent-s-biggest-lie-i-searched-the-web" ] }, { "slug": "cohere", "name": "Cohere", "url": "https://ai.engineer/orgs/cohere", "summary": "Enterprise agents, retrieval-augmented generation, and tool use connect Cohere's recorded AIE appearances. Shaan Desai examines agent reliability, Sandra Kublik presents the Command R model family and practical retrieval challenges, and Vivek Muppalla discusses model capabilities and enterprise deployment choices. These talks offer complementary starting points for implementation details and infrastructure decisions.", "speakerSlugs": [ "shaan-desai", "sandra-kublik", "vivek-muppalla" ], "talkSlugs": [ "building-enterprise-llm-agents-that-work", "building-sota-open-weights-tool-use-the-command-r-family", "cohere-for-vps-of-ai" ] }, { "slug": "databricks", "name": "Databricks", "url": "https://ai.engineer/orgs/databricks", "summary": "Durable AI architecture and reliable enterprise agents connect the recorded AIE appearances of Omar Khattab and Sandipan Bhaumik associated with Databricks. Khattab addresses engineering through model and optimization changes, while Bhaumik examines distributed agent coordination and the evaluation, governance, and operational preparation needed to move beyond demos.", "speakerSlugs": [ "sandipan-bhaumik", "omar-khattab" ], "talkSlugs": [ "from-chaos-to-choreography-multi-agent-orchestration-patterns-that-actually-work-sandipan-bhaumi", "on-engineering-ai-systems-that-endure-the-bitter-lesson", "the-production-ai-playbook-deploying-agents-at-enterprise-scale" ] }, { "slug": "elastic", "name": "Elastic", "url": "https://ai.engineer/orgs/elastic", "summary": "Elastic's recorded AIE appearances cover retrieval foundations with Philipp Krenn and agentic search with Leonie Monigatti. Their workshops connect keyword, vector, and hybrid search to the practical work of gathering context for agents, while Krenn's benchmarking talk examines how engineers can evaluate search performance against the workloads their applications actually use.", "speakerSlugs": [ "leonie-monigatti", "philipp-krenn" ], "talkSlugs": [ "agentic-search-for-context-engineering", "information-retrieval-from-the-ground-up", "vector-search-benchmark-eting" ] }, { "slug": "every", "name": "Every", "url": "https://ai.engineer/orgs/every", "summary": null, "speakerSlugs": [ "alex-duffy", "dan-shipper", "kieran-klaassen" ], "talkSlugs": [ "benchmarks-are-memes-how-what-we-measure-shapes-ai-and-us", "dispatch-from-the-future-building-an-ai-native-company-dan-shipper-every-ai-i", "the-era-of-compound-engineering" ] }, { "slug": "evil-martians", "name": "Evil Martians", "url": "https://ai.engineer/orgs/evil-martians", "summary": "Proactive AI interfaces, developer-tool product strategy, and founder-led distribution connect Evil Martians' recorded AIE appearances. Arthur Objartel demonstrates assistance inside existing workflows using Tegon, while Victoria Melnikova examines Bolt.new's product direction and founders' role in reaching customers. Together, these talks connect interface decisions with technical differentiation and go-to-market choices.", "speakerSlugs": [ "arthur-objartel", "victoria-melnikova" ], "talkSlugs": [ "don-t-just-slap-on-a-chatbot-building-ai-that-works-before-you-ask", "gtm-is-you-victoria-melnikova-evil-martians", "why-bolt-new-won-and-most-devtools-ai-pivots-failed-victoria-melnikova" ] }, { "slug": "hex", "name": "Hex", "url": "https://ai.engineer/orgs/hex", "summary": "AI engineering team design and practical product delivery anchor Hex's recorded AIE appearances with Dr Bryan Bischof, also credited as Bryan Bischof. His hiring talk examines the skills and collaboration needed to build a team; shared discussions address integrating AI into existing software and the operational demands of taking LLM applications beyond demos.", "speakerSlugs": [ "dr-bryan-bischof", "bryan-bischof" ], "talkSlugs": [ "hiring-building-an-ai-engineering-team", "the-ai-pivot-with-chris-white-of-prefect-bryan-bischof-of-hex", "what-we-learned-from-a-year-of-building-with-llms" ] }, { "slug": "hypermode", "name": "Hypermode", "url": "https://ai.engineer/orgs/hypermode", "summary": "Incremental AI application development and graph retrieval connect Hypermode's recorded AIE appearances. Kevin Van Gundy introduces an AI runtime and developer tools, Ryan Fox-Tyler demonstrates application building through GitHub Issues triage, and William Lyon presents a hybrid knowledge-graph architecture for GraphRAG, extending the discussion to graph-powered recommendations and agents.", "speakerSlugs": [ "ryan-fox-tyler", "kevin-van-gundy", "william-lyon" ], "talkSlugs": [ "git-push-get-an-ai-api", "hypermode-launch", "the-knowledge-graph-mullet-trimming-graphrag-complexity" ] }, { "slug": "ibm", "name": "IBM", "url": "https://ai.engineer/orgs/ibm", "summary": "Agent reliability and retrieval infrastructure connect IBM’s recorded AIE appearances: Tejas Kumar demonstrates browser-agent harnesses, Roy Derks examines tool calling, and Phil Nash introduces OpenRAG. These talks focus on engineering around language models, from checking actions and designing tool interfaces to processing documents for retrieval and configuring a deployable RAG stack.", "speakerSlugs": [ "tejas-kumar", "phil-nash", "roy-derks" ], "talkSlugs": [ "ai-agent-harnesses", "openrag-an-open-source-stack-for-rag-phil-nash", "tool-calling-is-not-just-plumbing-for-ai-agents" ] }, { "slug": "keycard", "name": "Keycard", "url": "https://ai.engineer/orgs/keycard", "summary": "Agent authorization anchors Keycard's recorded AIE appearances: Jared Hanson explains OAuth delegation, while Kim Maida examines credentials for individual tool calls. Allie Howe moderates a joint debate featuring Ian Livingstone on whether disciplined coding-agent loops deliver practical autonomy, extending the reading path from access controls to questions of engineering reliability.", "speakerSlugs": [ "jared-hanson", "kim-maida", "allie-howe", "ian-livingstone" ], "talkSlugs": [ "how-to-secure-agents-using-oauth", "it-s-10pm-do-you-know-where-your-agents-are", "the-great-loops-debate-dex-horthy-geoff-huntley-ian-livingstone-greg-pstrucha-moderated-by-allie" ] }, { "slug": "meta", "name": "Meta", "url": "https://ai.engineer/orgs/meta", "summary": "Agent reliability and AI-assisted game creation connect Meta’s recorded AIE appearances. Nishant Gupta examines infrastructure for stochastic agents and evaluation of complete production workflows. Danielle An and David Hoe demonstrate an audience-controlled game, considering how generative tools change creation while artistic cohesion, gameplay, and human taste remain central to the result.", "speakerSlugs": [ "nishant-gupta", "danielle-an", "david-hoe" ], "talkSlugs": [ "building-deterministic-infrastructure-for-non-deterministic-ai-agents", "production-evals-for-agentic-ai-systems", "think-you-can-build-a-game-with-ai-think-again-the-new-games-are-just-being-invented" ] }, { "slug": "netflix", "name": "Netflix", "url": "https://ai.engineer/orgs/netflix", "summary": "Netflix-affiliated AIE appearances connect AI-assisted software engineering with recommendation modeling. Jake Nations examines how generated code can outpace human understanding, Rajat Shah describes agents that investigate production CPU costs, and Yesu Feng presents shared personalization models. Together, these talks offer distinct perspectives on architectural complexity, performance optimization, and reusable machine-learning infrastructure.", "speakerSlugs": [ "rajat-shah", "jake-nations", "yesu-feng" ], "talkSlugs": [ "ai-agents-for-performance-ship-faster-pay-less", "infinite-software-crisis", "one-model-to-rule-recommendations-netflix-s-big-bet" ] }, { "slug": "openpipe", "name": "OpenPipe", "url": "https://ai.engineer/orgs/openpipe", "summary": "At AI Engineer, Kyle Corbitt’s OpenPipe appearances cover the move from prompting to task-specific models and reinforcement learning for reliable agents. His talks examine when specialization is justified and how an email assistant was trained; a joint session with Mustafa Ali of Method Financial discusses financial-agent workloads and production economics.", "speakerSlugs": [ "kyle-corbitt" ], "talkSlugs": [ "how-to-train-your-agent-building-reliable-agents-with-rl", "how-we-scaled-500m-ai-agents-in-production-with-2-engineers", "the-genai-maturity-curve-or-you-probably-don-t-need-fine-tuning" ] }, { "slug": "openrouter", "name": "OpenRouter", "url": "https://ai.engineer/orgs/openrouter", "summary": "Alex Atallah's AIE appearances cover OpenRouter's evolution from a browser extension into a multi-provider model marketplace and its model-routing roadmap. A joint panel with participants from NVIDIA and Cognition broadens the subject to how routing decisions balance model capability, application quality, inference cost, and local hardware utilization.", "speakerSlugs": [ "alex-atallah" ], "talkSlugs": [ "fun-stories-from-building-openrouter-and-where-all-this-is-going", "the-next-unicorns-7-top-ai-startups-from-the-hf0-residency", "the-state-of-model-routing-nvidia-cognition-openrouter" ] }, { "slug": "orb", "name": "Orb", "url": "https://ai.engineer/orgs/orb", "summary": "AI-agent pricing connects Orb's recorded AIE appearances by Alvaro Morales and Kshitij Grover. Morales examines monetization under volatile inference costs and demands for measurable ROI; Grover considers how customer value, purchasing behavior, and infrastructure economics shape prices. Together, these talks connect commercial choices with the engineering needed to revise and collect revenue.", "speakerSlugs": [ "alvaro-morales", "kshitij-grover" ], "talkSlugs": [ "monetizing-ai-alvaro-morales-orb", "revenue-engineering-how-to-price-and-reprice-your-ai-product", "the-price-of-intelligence-ai-agent-pricing-in-2025" ] }, { "slug": "parlance-labs", "name": "Parlance Labs", "url": "https://ai.engineer/orgs/parlance-labs", "summary": "Domain-specific LLM evaluations and the organizational work of shipping AI connect Hamel Husain's recorded appearances for Parlance Labs at AIE. Joint sessions examine Rechat's real-estate assistant, lessons from building LLM applications, and a satirical account of enterprise strategy failures, linking technical reliability to practical decisions about teams, costs, and business outcomes.", "speakerSlugs": [ "hamel-husain" ], "talkSlugs": [ "how-to-build-an-ai-strategy-that-fails", "how-to-construct-domain-specific-llm-evaluation-systems", "what-we-learned-from-a-year-of-building-with-llms" ] }, { "slug": "poolside", "name": "Poolside", "url": "https://ai.engineer/orgs/poolside", "summary": "Software-engineering models and agent verification connect Poolside’s recorded AIE appearances. Jason Warner and Eiso Kant present reinforcement learning as a path toward AGI; Marah Abdin and Robert McHardy examine synthetic data and pre-training failures; Johan Lajili describes application feedback that helps coding agents inspect and verify their work.", "speakerSlugs": [ "eiso-kant", "jason-warner", "marah-abdin", "robert-mchardy", "johan-lajili" ], "talkSlugs": [ "agi-the-path-forward", "the-messy-reality-of-scale-synthetic-data-and-pre-training", "your-agent-is-blindfolded" ] }, { "slug": "raindrop", "name": "Raindrop", "url": "https://ai.engineer/orgs/raindrop", "summary": "Production agent reliability connects Raindrop's AIE appearances by Ben Hylak, Danny Gollapalli, and Zubin Koticha. The sessions cover learning from user behavior, detecting failures in live systems, and verifying repairs. Hylak's reliability-floor argument sits alongside a practical observability workshop and a joint discussion with Oleve's Sid Bendre about dependable AI products.", "speakerSlugs": [ "ben-hylak", "danny-gollapalli", "zubin-koticha" ], "talkSlugs": [ "building-ai-products-that-actually-work", "designing-agents-the-floor-is-the-frontier", "everything-you-need-to-know-about-agent-observability" ] }, { "slug": "red-hat", "name": "Red Hat", "url": "https://ai.engineer/orgs/red-hat", "summary": "Portable agents, production evaluation, and document preparation connect Red Hat's recorded AIE appearances. Sally Ann O'Malley demonstrates containerized OpenClaw agents, Taylor Jordan Smith examines LLM evaluation beyond leaderboard accuracy, and Cedric Clyburn shows how Docling converts enterprise documents into structured inputs for retrieval and agent workflows.", "speakerSlugs": [ "sally-ann-o-malley", "taylor-jordan-smith", "cedric-clyburn" ], "talkSlugs": [ "lobster-trap-openclaw-in-containers-from-local-to-k8s-and-back", "strategies-for-llm-evals-guidellm-lm-eval-harness-openai-evals-workshop-taylor-jordan-smith", "structuring-the-unstructured-advanced-document-parsing-for-ai-workflows" ] }, { "slug": "roboflow", "name": "Roboflow", "url": "https://ai.engineer/orgs/roboflow", "summary": "Computer vision architecture and deployment connect Roboflow’s AIE appearances: Isaac Robinson explains the rise of vision transformers, while Peter Robicheaux examines visual reasoning and edge constraints. Joseph Nelson participates in a joint discussion of local AI adoption, extending the reading path from model design and evaluation to inference infrastructure and usability.", "speakerSlugs": [ "isaac-robinson", "joseph-nelson", "peter-robicheaux" ], "talkSlugs": [ "how-transformers-finally-ate-vision", "state-of-the-union-why-local-why-now", "vision-ai-in-2025-peter-robicheaux-roboflow" ] }, { "slug": "sierra", "name": "Sierra", "url": "https://ai.engineer/orgs/sierra", "summary": "Customer-facing agents connect Sierra’s AIE appearances: Clay Bavor discusses AI architecture with Alessio Fanelli, Zack Reneau-Wedeen examines reliable conversational-agent development, and Natalie Meurer traces forward-deployed engineering. Together, these sessions offer a reading path through agent design, operational testing, and the engineering roles responsible for customer outcomes.", "speakerSlugs": [ "clay-bavor", "zack-reneau-wedeen", "natalie-meurer" ], "talkSlugs": [ "rise-of-the-ai-architect-clay-bavor-and-alessio-fanelli", "the-agent-development-life-cycle", "the-dirty-secret-of-forward-deployed-engineering" ] }, { "slug": "sonar", "name": "Sonar", "url": "https://ai.engineer/orgs/sonar", "summary": "Verification of agent-generated software connects Sonar's recorded AIE appearances by Prasenjit Sarkar, Anirban Chatterjee, and Tariq Shaukat. Their talks examine enterprise code quality, the limits of human review, and engineering workflows that carry agent output through automated checks and remediation, linking coding productivity to the ongoing work of maintaining software.", "speakerSlugs": [ "prasenjit-sarkar", "anirban-chatterjee", "tariq-shaukat" ], "talkSlugs": [ "can-llms-generate-enterprise-quality-code-prasenjit-sarkar-sonar", "guide-verify-solve-the-engineering-discipline-agentic-development-demands", "in-the-land-of-ai-agents-the-verifiers-are-king" ] }, { "slug": "sourcegraph-amp", "name": "Sourcegraph/Amp", "url": "https://ai.engineer/orgs/sourcegraph-amp", "summary": null, "speakerSlugs": [ "steve-yegge", "beyang-liu" ], "talkSlugs": [ "2026-the-year-the-ide-died", "amp-code-next-generation-ai-coding", "the-emerging-skillset-of-wielding-coding-agents" ] }, { "slug": "supabase", "name": "Supabase", "url": "https://ai.engineer/orgs/supabase", "summary": "Supabase’s recorded AIE appearances connect PostgreSQL infrastructure with the agents that use it. Paul Copplestone presents pgvector for embedding-powered applications, while Pedro Rodrigues examines product-specific skills for MCP tools and demonstrates how to refine agent behavior. Together, these sessions cover vector search, database safety, and practical evaluation of coding workflows.", "speakerSlugs": [ "pedro-rodrigues", "paul-copplestone" ], "talkSlugs": [ "combine-skills-and-mcp-to-close-the-context-gap", "skill-issue-how-we-used-ai-to-make-agents-actually-good-at-supabase", "supabase-vector-the-postgres-vector-database" ] }, { "slug": "the-pragmatic-engineer", "name": "The Pragmatic Engineer", "url": "https://ai.engineer/orgs/the-pragmatic-engineer", "summary": "Gergely Orosz’s AIE appearances connect AI coding tools with questions about engineering performance, product judgment, and infrastructure economics. His conversations with swyx, Tuomas Artman, and Simon Eskildsen move between organizational pressure to adopt AI, the discipline of building useful products, and the practical costs of running software systems.", "speakerSlugs": [ "gergely-orosz" ], "talkSlugs": [ "building-redacted-username-gergely-orosz-simon-eskildsen", "software-engineering-ai", "taste-craft-a-conversation-with-tuomas-artman-cto-of-linear-and-gergely-orosz-of-the-pragmatic-e" ] }, { "slug": "towards-ai", "name": "Towards AI", "url": "https://ai.engineer/orgs/towards-ai", "summary": "Research agents, technical writing, and persistent context connect Towards AI’s AIE appearances featuring Louis-François Bouchard, Samridhi Vaid, and Omar Solano. Their recorded workshops examine an AI tutor’s conversational context, while joint sessions with Paul Iusztin discuss research-to-writing workflows and turning personal knowledge into reusable agent memory.", "speakerSlugs": [ "louis-francois-bouchard", "samridhi-vaid", "omar-solano" ], "talkSlugs": [ "build-your-own-deep-research-agent-technical-writer", "context-engineering-in-2026-compaction-memory-cost", "turn-10-994-notes-into-your-agents-memory" ] }, { "slug": "unblocked", "name": "Unblocked", "url": "https://ai.engineer/orgs/unblocked", "summary": "At AI Engineer, Jeff Ng, Peter Werry, and Brandon Waselnuk examine how coding agents use organizational knowledge to make engineering decisions. Their recorded talks connect missed incident history, retrieval failures, and context assembly to practical workflows including ticket enrichment, pull-request review, and a Zendesk integration.", "speakerSlugs": [ "jeff-ng", "peter-werry", "brandon-waselnuk" ], "talkSlugs": [ "building-agents-is-trivial-now-context-is-the-next-frontier", "mergeable-by-default-building-the-context-engine-to-save-time-and-tokens", "stop-babysitting-your-agents-building-a-context-engine-for-mergeable-code" ] }, { "slug": "vectara", "name": "Vectara", "url": "https://ai.engineer/orgs/vectara", "summary": "Enterprise RAG connects Ofer Mendelevitch's recorded Vectara appearances at AI Engineer: building the infrastructure, assessing its outputs, and extending retrieval into autonomous research. The talks cover operational responsibilities, evaluation without reference answers, and applications over private enterprise data, offering starting points for readers considering how to build, test, or apply RAG.", "speakerSlugs": [ "ofer-mendelevitch" ], "talkSlugs": [ "enterprise-deep-research-the-next-killer-app-for-enterprise-ai-ofer-mendelevitch-vectara", "open-rag-eval-rag-evaluation-without-golden-answers", "the-hidden-costs-of-building-your-own-rag-stack-ofer-mendelevitch-vectara" ] }, { "slug": "zapier", "name": "Zapier", "url": "https://ai.engineer/orgs/zapier", "summary": "Zapier’s recorded AIE appearances connect AI product evaluation with the practical work of diagnosing and fixing failures. Olmo Maldonado joins Braintrust’s Ankur Goyal to discuss AI Zap Builder and Zapier Copilot; Rafal Wilinski and Vitor Balocco examine agent evaluation, while Lisa Orr describes support staff helping ship integration fixes under engineering review.", "speakerSlugs": [ "olmo-maldonado", "rafal-wilinski", "vitor-balocco", "lisa-orr" ], "talkSlugs": [ "how-zapier-builds-ai-products-and-features-with-the-help-of-braintrust", "turning-fails-into-features-zapier-s-hard-won-eval-lessons", "your-support-team-should-ship-code" ] }, { "slug": "zed", "name": "Zed", "url": "https://ai.engineer/orgs/zed", "summary": "Zed’s AIE appearances address how coding agents connect to editors, how their changes are tested, and how edit predictions reach production. Bennet Fenner demonstrates an ACP-compatible agent, Nathan Sobo examines stochastic evaluations alongside deterministic CI, and Ben Kunkle explains Zeta2’s training and latency constraints. Together, these talks connect implementation choices with evaluation and operation.", "speakerSlugs": [ "bennet-fenner", "nathan-sobo", "ben-kunkle" ], "talkSlugs": [ "building-an-acp-compatible-agent-live-bennet-fenner-zed", "ci-in-the-era-of-ai-from-unit-tests-to-stochastic-evals", "how-we-built-zeta2-training-an-edit-prediction-model-in-production-ben-kunkle-zed" ] }, { "slug": "zep", "name": "Zep", "url": "https://ai.engineer/orgs/zep", "summary": "Daniel Chalef's AIE appearances connect agent memory with domain relevance, changing facts, and source lineage. His talks present Graphiti through a finance-coach demonstration and a healthcare provenance example, while his participation in a shared memory workshop places temporal graphs alongside other approaches to durable agent memory.", "speakerSlugs": [ "daniel-chalef" ], "talkSlugs": [ "citation-needed-provenance-for-llm-built-knowledge-graphs", "memory-masterclass-make-your-ai-agents-remember-what-they-do-mark-bain-aius", "stop-using-rag-as-memory" ] }, { "slug": "abundant", "name": "Abundant", "url": "https://ai.engineer/orgs/abundant", "summary": null, "speakerSlugs": [ "jesse-hu", "rishi-desai" ], "talkSlugs": [ "agents-are-robots-too-what-self-driving-taught-me-about-building-agents-jesse-hu-abundant", "swe-marathon-evaluating-coding-agents-at-billion-token-scale-rishi-desai-abundant-ai" ] }, { "slug": "agentic-ai-foundation", "name": "Agentic AI Foundation", "url": "https://ai.engineer/orgs/agentic-ai-foundation", "summary": "At AI Engineer World's Fair 2026, Angie Jones examined how to build dependable agentic systems and move engineering teams toward agent-driven software delivery. Her talks connect a house-hunting agent's design with an account of organizational change at Block, giving readers both implementation-level and organizational perspectives on delegating software work.", "speakerSlugs": [ "angie-jones" ], "talkSlugs": [ "build-systems-not-code", "building-an-autonomous-engineering-org" ] }, { "slug": "agentuity", "name": "Agentuity", "url": "https://ai.engineer/orgs/agentuity", "summary": "Rick Blalock's appearances at AI Engineer World's Fair 2025 connect the deployment demands of long-running agents with the organizational changes he argues they require. His talks cover serverless constraints and Agentuity deployment tooling, then consider what changes when humans orchestrate agents across products and everyday operations.", "speakerSlugs": [ "rick-blalock" ], "talkSlugs": [ "conquering-agent-chaos", "the-agent-native-company" ] }, { "slug": "ai21-labs", "name": "AI21 Labs", "url": "https://ai.engineer/orgs/ai21-labs", "summary": "RAG evaluation and agent planning anchor AI21 Labs' recorded appearances at AI Engineer World's Fair 2025. Yuval Belfer and Niv Granot examine why retrieval benchmarks miss questions requiring aggregation across documents; Belfer also discusses how agents can handle complex instructions through planning and adaptation, while keeping architecture as simple as the task allows.", "speakerSlugs": [ "yuval-belfer", "niv-granot" ], "talkSlugs": [ "rag-evaluation-is-broken-here-s-why-and-how-to-fix-it", "real-ai-agents-need-planning-not-just-prompting" ] }, { "slug": "alithea-bio", "name": "Alithea Bio", "url": "https://ai.engineer/orgs/alithea-bio", "summary": "At AI Engineer World's Fair 2026, Armanas Povilionis presented Froglet, an open agent-to-agent protocol for coordinating scientific and commercial services across organizational boundaries. The presentation combines service discovery with demonstrations of node interaction, Docker deployment, and provider-consumer roles.", "speakerSlugs": [ "armanas-povilionis" ], "talkSlugs": [ "agents-need-receipts-not-more-tool-calls", "agents-need-receipts-not-more-tool-calls-505635" ] }, { "slug": "alixpartners", "name": "AlixPartners", "url": "https://ai.engineer/orgs/alixpartners", "summary": "At AI Engineer, Kevin Madura and Mo Bhasin discuss generative AI for professional-services document analysis and the gap between enterprise enthusiasm and realized productivity. Madura's DSPy workshop adds an implementation perspective: building testable Python programs with structured outputs, document processing, and measured optimization instead of relying on manually tuned prompts.", "speakerSlugs": [ "kevin-madura", "mo-bhasin" ], "talkSlugs": [ "dspy-the-end-of-prompt-engineering", "the-billable-hour-is-dead-long-live-the-billable-hour" ] }, { "slug": "arcee-ai", "name": "Arcee AI", "url": "https://ai.engineer/orgs/arcee-ai", "summary": "Reasoning-oriented pre-training and control over local models connect Arcee AI’s recorded AIE appearances. Varun Singh argues that downstream reinforcement learning increasingly shapes base-model training data. Lucas Atkins joins a panel discussing why builders seek ownership, privacy, and durable access through local and open models, alongside representatives of NVIDIA and Prime Intellect.", "speakerSlugs": [ "lucas-atkins", "varun-singh" ], "talkSlugs": [ "local-models-trust-control-optimization", "the-base-model-is-dead" ] }, { "slug": "bespoke-labs", "name": "Bespoke Labs", "url": "https://ai.engineer/orgs/bespoke-labs", "summary": "Reasoning datasets and agent training environments connect Bespoke Labs’ recorded AIE appearances by Ryan Marten and Mahesh Sathiamoorthy. Marten presents OpenThoughts recipes for supervised fine-tuning across mathematics, coding, and science; Sathiamoorthy examines synthetic data and reinforcement-learning environments as bottlenecks in training reliable autonomous agents, extending the discussion into deployment infrastructure.", "speakerSlugs": [ "mahesh-sathiamoorthy", "ryan-marten" ], "talkSlugs": [ "data-and-environment-curation-for-post-training-llms", "openthoughts-data-recipes-for-reasoning-models" ] }, { "slug": "bismuth", "name": "Bismuth", "url": "https://ai.engineer/orgs/bismuth", "summary": "At AI Engineer World's Fair 2025, Ian Butler and Nick Gregory presented Bismuth's SM-100 benchmark to examine whether coding agents find genuine software defects. Their recorded appearances connect maintenance evaluation with practical coding workflows, covering misleading bug reports, developer alert fatigue, and checks that proposed repairs actually fix the problem.", "speakerSlugs": [ "ian-butler", "nick-gregory" ], "talkSlugs": [ "agents-reported-thousands-of-bugs-how-many-were-real-ian-butler-and-nick-gregory", "how-to-improve-your-vibe-coding-ian-butler" ] }, { "slug": "bloomberg", "name": "Bloomberg", "url": "https://ai.engineer/orgs/bloomberg", "summary": "At AI Engineer, Anju Kambadur discusses building agents for financial research analysts, while Lei Zhang examines AI deployment across Bloomberg’s engineering organization. Their talks connect agent reliability with practical delivery constraints: factuality, operational safeguards, verification, and the bottlenecks that remain when coding assistants make individual programming tasks faster.", "speakerSlugs": [ "anju-kambadur", "lei-zhang" ], "talkSlugs": [ "challenges-to-scaling-agents-for-generative-ai-products", "what-we-learned-deploying-ai-within-bloomberg-s-engineering-organization" ] }, { "slug": "browserbase", "name": "Browserbase", "url": "https://ai.engineer/orgs/browserbase", "summary": "At AI Engineer, Paul Klein IV presented browsers as a practical way for agents to work with websites lacking dedicated interfaces. His recorded talks connect browser-agent development with the engineering needed to operate reliably, moving from interaction methods and demonstrations to harnesses, reusable skills, memory, and infrastructure.", "speakerSlugs": [ "paul-klein-iv" ], "talkSlugs": [ "bringing-agents-onto-the-world-wide-web", "the-web-browser-is-all-you-need" ] }, { "slug": "cartesia", "name": "Cartesia", "url": "https://ai.engineer/orgs/cartesia", "summary": "Real-time multimodal intelligence connects Cartesia's recorded AIE appearances: Karan Goel examines architectures for responsive streaming systems, while Arjun Desai discusses enterprise voice AI with AWS's Rohit Talluri. Together, the sessions connect model design with deployment concerns, from persistent internal memory and low-power devices to latency across speech and language-model pipelines.", "speakerSlugs": [ "arjun-desai", "karan-goel" ], "talkSlugs": [ "serving-voice-ai-at-scale-arjun-desai-cartesia-rohit-talluri-aws", "state-space-models-for-realtime-multimodal-intelligence" ] }, { "slug": "catio", "name": "Catio", "url": "https://ai.engineer/orgs/catio", "summary": "Architecture copilots connect Catio’s recorded AIE appearances: Boris Bogatin and Toufic Boubez discuss improving technology decisions beyond coding productivity, while Iman Makaremi examines grounded reasoning for cloud architecture. Together, the talks connect enterprise priorities and visibility into existing systems with the technical design of agents that produce architectural recommendations.", "speakerSlugs": [ "boris-bogatin", "toufic-boubez", "iman-makaremi" ], "talkSlugs": [ "ai-copilots-for-tech-architecture-the-highest-roi-use-case-you-re-not-building-boris-bogatin-and", "grounded-reasoning-systems-for-cloud-architecture" ] }, { "slug": "cerebras", "name": "Cerebras", "url": "https://ai.engineer/orgs/cerebras", "summary": "Fast inference connects Cerebras’s recorded AIE appearances: Sarah Chieng examines how coding speed can accelerate productivity and technical debt, while Daniel Kim and Daria Soboleva introduce efficient model scaling and collaborative inference. Together, these sessions connect inference architecture with practical application building and the human work of directing coding agents.", "speakerSlugs": [ "sarah-chieng", "daniel-kim", "daria-soboleva" ], "talkSlugs": [ "fast-models-need-slow-developers", "from-mixture-of-experts-to-mixture-of-agents-with-super-fast-inference" ] }, { "slug": "character-ai", "name": "Character.ai", "url": "https://ai.engineer/orgs/character-ai", "summary": "Generated-video evaluation and multimodal training data connect Character.ai's recorded AIE appearances. Maor Bril describes methods for detecting temporal, visual, and audio failures in generated video. In a joint session with Chang She of LanceDB, Noah Shpak discusses the infrastructure used to inspect, construct, and refine datasets for foundation-model training.", "speakerSlugs": [ "maor-bril", "noah-shpak" ], "talkSlugs": [ "evaling-video-slop", "the-hierarchy-of-needs-for-training-dataset-development" ] }, { "slug": "chroma", "name": "Chroma", "url": "https://ai.engineer/orgs/chroma", "summary": "Retrieval quality connects Chroma’s recorded AIE appearances by Jeff Huber and Anton Troynikov. Huber examines how application queries expose differences between embedding models, while Troynikov discusses the retrieval and memory infrastructure needed for production RAG. Together, the talks connect measurement on real workloads with choices about how information reaches a model.", "speakerSlugs": [ "jeff-huber", "anton-troynikov" ], "talkSlugs": [ "how-to-look-at-your-data-what-to-look-for-how-to-measure", "retrieval-augmented-generation-in-the-wild" ] }, { "slug": "comet-ml-openclaw", "name": "Comet ML / OpenClaw", "url": "https://ai.engineer/orgs/comet-ml-openclaw", "summary": "Vincent Koc's appearances at AI Engineer Europe 2026 address supervising parallel coding agents and evaluating agents as their behavior changes. Drawing on OpenClaw development, the talks connect engineering judgment with ongoing feedback, presenting software delivery and evaluation as work that requires continued oversight as agents gain more responsibility.", "speakerSlugs": [ "vincent-koc" ], "talkSlugs": [ "dark-factory-openclaw-ships-faster-than-you-can-read-the-diff", "malleable-evals-why-are-we-still-evaluating-adaptive-systems-with-static-tests" ] }, { "slug": "contextual-ai", "name": "Contextual AI", "url": "https://ai.engineer/orgs/contextual-ai", "summary": "Contextual AI’s recorded AIE appearances address specialized RAG systems and practical agent building. Douwe Kiela connects enterprise value to retrieval quality, proprietary context, and production design, while Nina Lopatina and Rajiv Shah demonstrate a managed workflow for ingesting documents, creating agents, and evaluating their responses.", "speakerSlugs": [ "nina-lopatina", "rajiv-shah", "douwe-kiela" ], "talkSlugs": [ "forget-rag-pipelines-build-production-ready-ai-agents-in-15-minutes", "production-rag-agents" ] }, { "slug": "convex", "name": "Convex", "url": "https://ai.engineer/orgs/convex", "summary": "At AI Engineer World's Fair 2024, Tom Redman demonstrates a voice assistant that places telephone calls, while Jamie Turner explains reactive backend design for multistep generative AI applications. Convex's recorded appearances connect a concrete assistant build with the data synchronization and workflow primitives used to coordinate speech, model outputs, and application state.", "speakerSlugs": [ "tom-redman", "jamie-turner" ], "talkSlugs": [ "building-an-ai-assistant-that-makes-phone-calls", "convex-launch" ] }, { "slug": "dagger", "name": "DAgger", "url": "https://ai.engineer/orgs/dagger", "summary": "At AI Engineer World's Fair 2025, Solomon Hykes presents isolated environments for autonomous coding agents, while Kyle Penfound and Jeremy Adams - Casañas demonstrate building a software engineering agent with scoped tools. DAgger's recorded sessions connect control over agent execution with practical workflows for inspecting code, making changes, and running tests.", "speakerSlugs": [ "solomon-hykes", "kyle-penfound", "jeremy-adams-casanas" ], "talkSlugs": [ "containing-agent-chaos", "ship-agents-that-ship-a-hands-on-workshop-for-swe-agent-builders" ] }, { "slug": "datadog", "name": "Datadog", "url": "https://ai.engineer/orgs/datadog", "summary": "At AI Engineer, Diamond Bishop presents Bits AI agents that investigate operational alerts and generate application fixes, while Diane Lin examines why security agents reach contradictory decisions on identical inputs. These appearances connect operational automation with the problem of deciding when agents need clearer context or human judgment.", "speakerSlugs": [ "diamond-bishop", "diane-lin" ], "talkSlugs": [ "the-devops-engineer-who-never-sleeps", "why-your-agent-disagrees-with-itself-and-what-to-do-about-it" ] }, { "slug": "decoding-ai", "name": "Decoding AI", "url": "https://ai.engineer/orgs/decoding-ai", "summary": "Deep research, technical writing, and persistent agent memory connect Paul Iusztin's recorded AIE appearances for Decoding AI. In shared sessions with Louis-François Bouchard and, in the writing workshop, Samridhi Vaid, the demonstrations cover systems that turn research into written output and personal knowledge into reusable context for agents.", "speakerSlugs": [ "paul-iusztin" ], "talkSlugs": [ "build-your-own-deep-research-agent-technical-writer", "turn-10-994-notes-into-your-agents-memory" ] }, { "slug": "deepgram", "name": "Deepgram", "url": "https://ai.engineer/orgs/deepgram", "summary": "At AI Engineer World's Fair 2024, Scott Stephenson and Damien Murphy presented complementary views of responsive voice agents. Stephenson explained how speech recognition, language models, and speech synthesis work together, while Murphy's hands-on workshop covered building conversational agents with Node.js and browser JavaScript and taking them toward telephony deployment and scale.", "speakerSlugs": [ "damien-murphy", "scott-stephenson" ], "talkSlugs": [ "building-scaling-an-ai-agent-swarm-of-low-latency-real-time-voice-bots", "giving-a-voice-to-ai-agents" ] }, { "slug": "ebay", "name": "eBay", "url": "https://ai.engineer/orgs/ebay", "summary": "Agent safeguards and the human cost of reviewing AI-generated code connect Sachin Gupta's recorded eBay appearances at AI Engineer. His talks cover controlling agent behavior in production and estimating pull-request review burden, bringing operational safety and reviewer capacity into discussions of AI-assisted software delivery.", "speakerSlugs": [ "sachin-gupta" ], "talkSlugs": [ "agents-need-feature-flags", "reviewdebt-a-practical-framework-for-scoring-every-pull-request-sachin-gupta-ebay" ] }, { "slug": "exo-labs", "name": "EXO Labs", "url": "https://ai.engineer/orgs/exo-labs", "summary": "Running frontier language models on local hardware anchors Alex Cheema's recorded AIE appearances with EXO Labs. His workshop connects inference mechanics with ownership, privacy, and practical demonstrations; a joint panel broadens the discussion to local AI adoption, distributed inference, and the technical barriers that remain.", "speakerSlugs": [ "alex-cheema" ], "talkSlugs": [ "frontier-ai-at-home-literally", "state-of-the-union-why-local-why-now" ] }, { "slug": "factset", "name": "FactSet", "url": "https://ai.engineer/orgs/factset", "summary": "FactSet's recorded AIE appearances feature Yogendra Miraje on financial-research agents, from blueprint-guided planning to reusable skills executed by agent harnesses. His presentations connect an earnings-call preparation workflow with a later approach to research and report generation, showing how task structure and execution controls change as agent capabilities become reusable product features.", "speakerSlugs": [ "yogendra-miraje" ], "talkSlugs": [ "how-to-build-planning-agents-without-losing-control-yogendra-miraje-factset", "skills-are-new-features-building-skill-centric-harness-yogendra-miraje-factset" ] }, { "slug": "featherless-ai", "name": "Featherless.ai", "url": "https://ai.engineer/orgs/featherless-ai", "summary": "Open-weight model applications anchor Featherless.ai's recorded AIE presence: Eugene Cheah surveys companionship, creative writing, coding, retrieval, and agents, comparing individual and enterprise demand. A separate HF0 Residency showcase lists Eugene with Featherless.ai among startup presenters, providing broader startup context rather than another detailed account of model usage.", "speakerSlugs": [ "eugene", "eugene-cheah" ], "talkSlugs": [ "the-next-unicorns-7-top-ai-startups-from-the-hf0-residency", "wtf-do-people-use-open-models-for" ] }, { "slug": "fireworks-ai", "name": "Fireworks AI", "url": "https://ai.engineer/orgs/fireworks-ai", "summary": "At AI Engineer World's Fair 2024, Dmytro (Dima) Dzhulgakov discusses customizing open models to improve application latency, cost, and domain-specific quality. Substituting for scheduled speaker Lin Qiao, he covers fine-tuned Llama and Gemma models alongside the practical demands of production deployment and multimodal image-model inference.", "speakerSlugs": [ "dmytro-dima-dzhulgakov", "lin-qiao" ], "talkSlugs": [ "customized-production-ready-inference-with-open-source-models-dmytro-dima-dzhulgakov", "making-open-models-10x-faster-and-better-for-modern-application-innovation" ] }, { "slug": "freeplay", "name": "Freeplay", "url": "https://ai.engineer/orgs/freeplay", "summary": "Enterprise AI product strategy and production reliability connect Jeremy Silva’s recorded Freeplay appearances at AI Engineer World's Fair 2025. In joint discussions with Eliza Cabrera of Workday and Chris Hernandez of Chime, the sessions address how teams choose useful AI workflows and sustain them beyond promising prototypes.", "speakerSlugs": [ "jeremy-silva" ], "talkSlugs": [ "build-dynamic-products-and-stop-the-ai-sideshow", "the-build-operate-divide-bridging-product-vision-and-ai-operational-reality" ] }, { "slug": "google-labs", "name": "Google Labs", "url": "https://ai.engineer/orgs/google-labs", "summary": "Jules anchors the Google Labs talks collected here: Rustin Banks presents concurrent coding tasks, while Kath Korevec examines the attention required to initiate and coordinate agent work. These talks connect the practical work of reviewing parallel outputs with a proposed shift toward agents that understand repositories and suggest useful maintenance.", "speakerSlugs": [ "kath-korevec", "rustin-banks" ], "talkSlugs": [ "proactive-agents", "your-coding-agent-just-got-cloned-and-your-brain-isn-t-ready" ] }, { "slug": "graphite", "name": "Graphite", "url": "https://ai.engineer/orgs/graphite", "summary": "At AI Engineer World's Fair 2025, Graphite co-founder Tomas Reimers discusses AI code review and the work required to ship AI-generated code. His talks examine Diamond’s pull-request feedback and argue that faster coding increases pressure on testing, review, merging, and deployment, making those steps central to the development workflow.", "speakerSlugs": [ "tomas-reimers" ], "talkSlugs": [ "ai-powered-entomology-lessons-from-millions-of-ai-code-reviews", "don-t-get-one-shotted-use-ai-to-test-review-merge-and-deploy-code-tomas-reimers-graphite" ] }, { "slug": "hasura", "name": "Hasura", "url": "https://ai.engineer/orgs/hasura", "summary": "Hybrid retrieval and secure access to business data connect Hasura's recorded AIE appearances. Simrat Hanspal demonstrates e-commerce search combining vectors and structured filters, while Tanmai Gopal introduces Pacha DDN for connecting LLMs to live data and business logic. Their examples span product discovery and customer-email workflows.", "speakerSlugs": [ "simrat-hanspal", "tanmai-gopal" ], "talkSlugs": [ "building-efficient-hybrid-context-query-for-llm-grounding", "hasura-launch-realtime-data-connectivity-for-ai" ] }, { "slug": "intercom", "name": "Intercom", "url": "https://ai.engineer/orgs/intercom", "summary": "Intercom’s recorded AIE appearances connect AI-assisted software development with enterprise phone support. Brian Scanlan describes an effort to double engineering throughput through coding-agent adoption and measurement, while Peter Bar explains how Fin Voice reached an initial product in approximately 100 days, answering customer questions and escalating to humans when necessary.", "speakerSlugs": [ "brian-scanlan", "peter-bar" ], "talkSlugs": [ "how-building-with-ai-can-double-the-throughput-of-your-engineering-team", "shipping-an-enterprise-voice-ai-agent-in-100-days" ] }, { "slug": "intuit", "name": "Intuit", "url": "https://ai.engineer/orgs/intuit", "summary": "At AI Engineer, Jaspreet Singh describes personalized tax explanations in TurboTax, while Udi Menkes examines why general-purpose LLMs struggle with financial advice. Their recorded talks connect financial AI to domain-specific evidence: proprietary tax knowledge for answering taxpayer questions, and business trajectories for assessing the consequences of possible actions.", "speakerSlugs": [ "jaspreet-singh", "udi-menkes" ], "talkSlugs": [ "how-intuit-uses-llms-to-explain-taxes-to-millions-of-taxpayers", "why-off-the-shelf-ai-doesn-t-understand-money" ] }, { "slug": "kepler", "name": "Kepler", "url": "https://ai.engineer/orgs/kepler", "summary": "At AI Engineer World's Fair 2026, Vinoo Ganesh discusses verifiable financial AI, explaining why evaluations alone cannot establish trust in language-model outputs. A separate session connected to Kepler examines forward deployed engineering as product discovery, grounding delivery decisions in customers' actual problems and observed workflows.", "speakerSlugs": [ "vinoo-ganesh" ], "talkSlugs": [ "how-forward-deployed-engineering-is-done-redacted-username-kepler", "how-kepler-built-verifiable-ai-for-financial-services" ] }, { "slug": "lancedb", "name": "LanceDB", "url": "https://ai.engineer/orgs/lancedb", "summary": "LanceDB's recorded AIE appearances feature Chang She in joint discussions about enterprise legal RAG and multimodal training datasets. With Calvin Qi of Harvey and Noah Shpak of Character.ai, respectively, the sessions connect retrieval and model development requirements to data infrastructure choices, including ingestion, storage, dataset refinement, and access patterns.", "speakerSlugs": [ "chang-she" ], "talkSlugs": [ "scaling-enterprise-grade-rag-systems-lessons-from-the-legal-frontier", "the-hierarchy-of-needs-for-training-dataset-development" ] }, { "slug": "langbase", "name": "Langbase", "url": "https://ai.engineer/orgs/langbase", "summary": null, "speakerSlugs": [ "ahmad-awais" ], "talkSlugs": [ "developing-taste-in-coding-agents-applied-meta-neuro-symbolic-rl-ahmad-awais-command-code", "why-the-best-ai-agents-are-built-without-frameworks-primitives-over-frameworks" ] }, { "slug": "langfuse", "name": "Langfuse", "url": "https://ai.engineer/orgs/langfuse", "summary": null, "speakerSlugs": [ "marc-klingen", "annabell-schafer" ], "talkSlugs": [ "skill-issue-lessons-from-skilling-up-coding-agents-to-use-langfuse", "stop-burning-tokens-why-self-improvement-needs-domain-expertise-first-annabell-schafer-langfuse" ] }, { "slug": "linear", "name": "Linear", "url": "https://ai.engineer/orgs/linear", "summary": "At AI Engineer, Tom Moor describes Linear's progression from embedded AI features to coordinating human and AI teammates. Tuomas Artman's conversation with Gergely Orosz examines product judgment and craftsmanship in AI-assisted development. Together, these appearances connect the infrastructure for agent participation with the engineering habits used to maintain product quality.", "speakerSlugs": [ "tom-moor", "tuomas-artman" ], "talkSlugs": [ "building-the-platform-for-agent-coordination", "taste-craft-a-conversation-with-tuomas-artman-cto-of-linear-and-gergely-orosz-of-the-pragmatic-e" ] }, { "slug": "linkedin", "name": "LinkedIn", "url": "https://ai.engineer/orgs/linkedin", "summary": "Personalized ranking and generative AI infrastructure connect LinkedIn's recorded AIE appearances. Hamed Firooz and Maziar Sanjabi present 360Brew, a unified foundation-model approach to recommendation, while Xiaofeng Wang traces a platform's evolution from Collaborative Articles to retrieval-augmented coaching and the multi-agent Hiring Assistant. Together, the talks cover model development and application infrastructure.", "speakerSlugs": [ "hamed-firooz", "maziar-sanjabi", "xiaofeng-wang" ], "talkSlugs": [ "360brew-llm-based-personalized-ranking-and-recommendation-hamed-firooz-and-maziar-sanjabi-linked", "lessons-from-building-linkedin-s-genai-platform" ] }, { "slug": "liquid-ai", "name": "Liquid AI", "url": "https://ai.engineer/orgs/liquid-ai", "summary": "Maxime Labonne's AIE appearances for Liquid AI cover training small language models for constrained devices and adapting open models through post-training and merging. The talks connect architecture and inference choices with dataset construction, alignment, and evaluation, giving readers practical considerations for both building small models and customizing existing ones.", "speakerSlugs": [ "maxime-labonne" ], "talkSlugs": [ "everything-i-learned-training-frontier-small-models", "everything-you-need-to-know-about-finetuning-and-merging-llms" ] }, { "slug": "log10", "name": "Log10", "url": "https://ai.engineer/orgs/log10", "summary": "LLM evaluation connects Log10’s recorded AIE appearances: Niklas Nielsen demonstrates local testing with llmeval, while Arjun Bansal joins Echo AI’s Trey Doig to discuss enterprise conversation intelligence. The sessions move from organizing prompts and comparing model outputs to assessing customer-facing systems and supporting production throughput with domain-adapted models.", "speakerSlugs": [ "niklas-nielsen", "arjun-bansal" ], "talkSlugs": [ "best-practices-for-evaluating-large-language-model-applications-with-llmeval-niklas-nielsen", "what-it-actually-takes-to-deploy-genai-applications-to-enterprises" ] }, { "slug": "mastra", "name": "Mastra", "url": "https://ai.engineer/orgs/mastra", "summary": "Sam Bhagwat’s AIE appearances examine how agents combine with structured workflows and how coding harnesses might evolve into always-on personal agents. Across these talks, he moves from architectural composition to persistent operation, discussing supervisor agents, background execution, interactive steering, and durable memory as parts of increasingly capable agent systems.", "speakerSlugs": [ "sam-bhagwat" ], "talkSlugs": [ "agents-vs-workflows-why-not-both", "every-harness-will-become-a-claw" ] }, { "slug": "metr", "name": "METR", "url": "https://ai.engineer/orgs/metr", "summary": "At AI Engineer Code 2025, METR-affiliated speaker Joel Becker examined how AI task-completion horizons relate to productivity in real software development. His talk contrasted capability measurements with a randomized field study of experienced open-source developers, while his workshop considered adoption patterns and evaluation challenges in mature repositories.", "speakerSlugs": [ "joel-becker" ], "talkSlugs": [ "long-tasks-and-experienced-open-source-dev-productivity", "why-agent-hype-can-fall-short-of-reality-joel-becker-metr" ] }, { "slug": "microsoft-research", "name": "Microsoft Research", "url": "https://ai.engineer/orgs/microsoft-research", "summary": "At AI Engineer World's Fair 2025, Jonathan Larson examines structured memory for retrieval and repository-level code understanding, while Victor Dibia presents human-centered design for semi-autonomous multi-agent systems. Their recorded talks connect GraphRAG demonstrations and retrieval evaluation with practical questions about how people discover agent capabilities, monitor execution, and retain control.", "speakerSlugs": [ "jonathan-larson", "victor-dibia" ], "talkSlugs": [ "graphrag-methods-to-create-optimized-llm-context-windows-for-retrieval-jonathan-larson-microsoft", "ux-design-principles-for-semi-autonomous-multi-agent-systems" ] }, { "slug": "minimax", "name": "MiniMax", "url": "https://ai.engineer/orgs/minimax", "summary": "Olive Song's AIE appearances connect open-weight coding agents with the infrastructure needed to serve agent workloads. Her MiniMax-M2 presentation covers reinforcement learning, tool use, and robustness; a joint discussion with Together AI's Dan Fu examines MiniMax-M3, multimodal model serving, and the demands of long-horizon tasks.", "speakerSlugs": [ "olive-song" ], "talkSlugs": [ "agents-at-scale-inside-minimax-s-model-and-the-infrastructure-behind-it", "minimax-m2" ] }, { "slug": "mistral-ai", "name": "Mistral AI", "url": "https://ai.engineer/orgs/mistral-ai", "summary": "Mistral AI’s recorded AIE appearances cover language models and speech generation. Devendra Singh Chaplot surveys Mistral, Mixtral, and Codestral alongside training methods and deployment options, while Samuel Humeau uses Voxtral TTS to explain contemporary speech synthesis, demonstrate voice cloning and multilingual output, and examine how voice-agent components fit together.", "speakerSlugs": [ "devendra-chaplot", "devendra-singh-chaplot", "samuel-humeau" ], "talkSlugs": [ "decoding-mistral-ai-s-large-language-models", "why-tts-models-now-look-like-llms-samuel-humeau-mistral" ] }, { "slug": "morgan-stanley", "name": "Morgan Stanley", "url": "https://ai.engineer/orgs/morgan-stanley", "summary": "Autonomous research and agent learning connect Morgan Stanley’s recorded AIE appearances. Brendan Rappazzo presents AlphaLab, a multi-agent system that starts from datasets and natural-language objectives; Will Brown examines how reinforcement learning could move language-model systems beyond engineered workflows. Together, the talks offer perspectives on coordinating research experiments and training agents to use tools.", "speakerSlugs": [ "brendan-rappazzo", "will-brown" ], "talkSlugs": [ "alphalab-autonomous-multi-agent-research-across-optimization-domains-with-frontier-llms-brendan", "what-rl-means-for-agents" ] }, { "slug": "novartis", "name": "Novartis", "url": "https://ai.engineer/orgs/novartis", "summary": "Hubert Misztela's recorded AIE talks connect scientific discovery with the practical preparation needed for enterprise agents. In these Novartis-affiliated appearances, he examines how language models might connect findings across research domains and how organizations can prepare for increasingly autonomous systems through workflow mapping, operational context, and attention to ethics.", "speakerSlugs": [ "hubert-misztela" ], "talkSlugs": [ "agentic-enterprise-what-your-ceo-must-know-about-ai", "llm-scientific-reasoning-how-to-make-ai-capable-of-nobel-prize-discoveries" ] }, { "slug": "nubank", "name": "Nubank", "url": "https://ai.engineer/orgs/nubank", "summary": "Nubank’s recorded AIE appearances address customer-support agent evaluation and security for AI development tools. Aman Gupta joins Shreya Rajpal of Snowglobe to discuss simulation-based testing, while Lucas Palma presents a workflow for reviewing AI skills before developer distribution. Together, these sessions examine testing and review before agents and development tools reach users.", "speakerSlugs": [ "aman-gupta", "lucas-palma" ], "talkSlugs": [ "simulation-maxxing-how-nubank-ships-agents-20-faster-with-simulations", "we-vetted-2-000-ai-skills-before-they-reached-developers" ] }, { "slug": "oleve", "name": "Oleve", "url": "https://ai.engineer/orgs/oleve", "summary": "Consumer AI operations and product reliability connect Sid Bendre’s appearances at AI Engineer World's Fair 2025. His presentation covers building Quizard AI and Unstuck AI with a small team, while a joint discussion with Ben Hylak of Raindrop examines dependable products through iteration and observation of real-world behavior.", "speakerSlugs": [ "sid-bendre" ], "talkSlugs": [ "building-ai-products-that-actually-work", "the-new-lean-startup" ] }, { "slug": "ona", "name": "Ona", "url": "https://ai.engineer/orgs/ona", "summary": null, "speakerSlugs": [ "lou-bichard" ], "talkSlugs": [ "building-ciso-approved-agent-fleet-architecture", "the-missing-primitive-for-agent-swarms" ] }, { "slug": "osmantic", "name": "Osmantic", "url": "https://ai.engineer/orgs/osmantic", "summary": "At AI Engineer World's Fair 2026, Ahmad Osman of Osmantic examines how improving open models could bring frontier-class capabilities to desktop workstations and consumer GPUs. His solo talk connects hardware requirements with local compute ownership, while a shared panel considers the deployment tools and technical challenges shaping local AI adoption.", "speakerSlugs": [ "ahmad-osman" ], "talkSlugs": [ "state-of-the-union-why-local-why-now", "the-desktop-frontier-ahmad-osman-osmantic" ] }, { "slug": "paperclip", "name": "Paperclip", "url": "https://ai.engineer/orgs/paperclip", "summary": "Dotta's AIE appearances examine how Paperclip organizes AI-agent work and how people determine whether that work is complete. The Europe demonstration covers delegated roles, shared context, and recurring routines; the World's Fair talk separates merging, deploying, and announcing into decisions with distinct evidence, authority, and risk ownership.", "speakerSlugs": [ "dotta" ], "talkSlugs": [ "paperclip-open-source-human-control-plane-for-ai-labor-dotta", "what-does-done-even-mean-agents-and-paperclip-s-liveness-model-dotta-paperclip" ] }, { "slug": "pi-labs", "name": "Pi Labs", "url": "https://ai.engineer/orgs/pi-labs", "summary": "David Karam's appearances for Pi Labs at AI Engineer World's Fair 2025 focus on evaluating LLM applications and improving retrieval-augmented generation through observed failures. The workshop turns application requirements into scoring workflows, while the retrieval talk examines how real queries guide successive changes to search, ranking, and orchestration.", "speakerSlugs": [ "david-karam" ], "talkSlugs": [ "building-metrics-that-actually-work-david-karam-pi-labs", "layering-every-technique-in-rag-one-query-at-a-time" ] }, { "slug": "pinterest", "name": "Pinterest", "url": "https://ai.engineer/orgs/pinterest", "summary": "Pinterest's AIE appearances address search relevance and failed Spark jobs. Han Wang and Mukuntha Narayanan describe LLM relevance classifiers enriched with visual captions and user signals; Drasko Profirovic presents Medic for Apache Spark. Together, these talks offer practical accounts of serving learned relevance models and building agents that diagnose distributed computing failures.", "speakerSlugs": [ "drasko-profirovic", "mukuntha-narayanan", "han-wang" ], "talkSlugs": [ "medic-for-apache-spark-first-aid-for-failing-jobs-drasko-profirovic-pinterest", "what-we-learned-from-using-llms-in-pinterest" ] }, { "slug": "posthog", "name": "PostHog", "url": "https://ai.engineer/orgs/posthog", "summary": "At AI Engineer Europe 2026, Danilo Campos examines failures in automated PostHog integration, while Joshua Snyder describes turning product signals into pull requests. These recorded talks connect practical coding-agent guidance with a broader development pipeline, giving readers complementary views of how agents receive context, investigate problems, and produce changes.", "speakerSlugs": [ "danilo-campos", "joshua-snyder" ], "talkSlugs": [ "llm-codegen-fails-and-how-to-stop-em", "self-driving-products-product-signals-to-pull-requests" ] }, { "slug": "prefect", "name": "Prefect", "url": "https://ai.engineer/orgs/prefect", "summary": "Prefect’s recorded AIE appearances connect AI product integration with tool design for agents. Chris White joins a discussion about established software companies investing in AI, including evaluation and rollout strategy. Jeremiah Lowin examines how MCP servers can serve agent workflows, covering focused tools, orchestration costs, and client behavior.", "speakerSlugs": [ "chris-white", "jeremiah-lowin" ], "talkSlugs": [ "the-ai-pivot-with-chris-white-of-prefect-bryan-bischof-of-hex", "your-mcp-server-is-bad-and-you-should-feel-bad" ] }, { "slug": "prompthub", "name": "PromptHub", "url": "https://ai.engineer/orgs/prompthub", "summary": "Dan Cleary’s AIE appearances connect practical prompt design with LLM product quality, from source grounding and multi-persona collaboration to techniques for reasoning models. Across the 2023 and 2025 sessions, he presents prompting as an early intervention before adding architectural complexity, while examining how examples and context can affect model performance.", "speakerSlugs": [ "dan-cleary" ], "talkSlugs": [ "prompt-engineering-tactics", "the-model-isn-t-wrong-you-re-just-bad-at-prompting" ] }, { "slug": "quotient", "name": "Quotient", "url": "https://ai.engineer/orgs/quotient", "summary": "AI search evaluation and RAG optimization frame Quotient's recorded AIE appearances with Julia Neagu and Deanna Emery. In joint sessions with Tavily and Qdrant speakers, they discuss testing search agents against changing web content and using systematic evaluation to guide retrieval improvements without unnecessary architectural complexity.", "speakerSlugs": [ "julia-neagu", "deanna-emery" ], "talkSlugs": [ "evaluating-ai-search-a-practical-framework-for-augmented-ai-systems", "navigating-rag-optimization-with-an-evaluation-driven-compass" ] }, { "slug": "ramp", "name": "Ramp", "url": "https://ai.engineer/orgs/ramp", "summary": "Ramp’s recorded AIE appearances connect agent architecture with enterprise engineering decisions. Rahul Sengottuvelu compares approaches to financial CSV conversion and considers how model capabilities could change software interfaces. Leo Mehr examines how to scope customer requests before building and where agents can assist an engineering workflow while retaining human judgment.", "speakerSlugs": [ "leo-mehr", "rahul-sengottuvelu" ], "talkSlugs": [ "how-forward-deployed-engineering-is-done-at-ramp", "scaffold-wisely" ] }, { "slug": "reflection-ai", "name": "Reflection AI", "url": "https://ai.engineer/orgs/reflection-ai", "summary": "Autonomous coding anchors Reflection AI’s recorded AIE appearances: Aakanksha Chowdhery connects language-model scaling, inference-time sampling, and reinforcement learning to software engineering. Tina participates in Frontier Feud, a shared game-show session where participants discuss predictions about frontier models, coding agents, and deployment alongside a contest about AI researchers and technical references.", "speakerSlugs": [ "tina", "aakanksha-chowdhery" ], "talkSlugs": [ "frontier-feud", "rl-for-autonomous-coding-aakanksha-chowdhery-reflection-ai" ] }, { "slug": "replit", "name": "Replit", "url": "https://ai.engineer/orgs/replit", "summary": "At AI Engineer, Amjad Masad and Michele Catasta present broadly available coding assistance and the engineering behind smaller code models. Catasta's later appearance examines autonomy for nontechnical users, shifting the focus from assistance inside development environments to agents that complete substantial software tasks without expert supervision.", "speakerSlugs": [ "amjad-masad", "michele-catasta" ], "talkSlugs": [ "building-ai-for-all", "the-3-pillars-of-autonomy-michele-catasta-replit" ] }, { "slug": "rexmore", "name": "Rexmore", "url": "https://ai.engineer/orgs/rexmore", "summary": "Rexmore's recorded AIE appearances address maintainable AI-assisted software development and agents that operate websites. At AI Engineer Code 2025, Corey J. Gallon presents a structured alternative to unplanned vibe coding; at AI Engineer World's Fair 2026, Corey Gallon presents chrome-agent and techniques for interacting with complex browser interfaces.", "speakerSlugs": [ "corey-j-gallon", "corey-gallon" ], "talkSlugs": [ "the-cure-for-the-vibe-coding-hangover", "the-dark-arts-of-web-automation-teaching-agents-to-use-websites-like-humans" ] }, { "slug": "root-signals", "name": "Root Signals", "url": "https://ai.engineer/orgs/root-signals", "summary": "Agent evaluation connects Root Signals' recorded AIE appearances through Ari Heljakka's talks on evaluation design and MCP-connected judges. He presents a map separating semantic quality from behavioral performance, then demonstrates how persistent evaluators can provide feedback on agent outputs, with examples spanning marketing copy and hotel reservations.", "speakerSlugs": [ "ari-heljakka" ], "talkSlugs": [ "agent-evals-finally-with-the-map", "will-agent-evaluation-via-mcp-stabilize-agent-networks" ] }, { "slug": "runpod", "name": "RunPod", "url": "https://ai.engineer/orgs/runpod", "summary": "At AI Engineer Europe 2026, Audry Hsu demonstrates two routes to cloud GPU execution: deploying an LLM endpoint through RunPod’s console and sending Python functions to remote workers with Flash. The sessions connect infrastructure choices with practical development workflows, from configuring endpoint workers to keeping surrounding application code local.", "speakerSlugs": [ "audry-hsu" ], "talkSlugs": [ "gpu-cloud-deployment-without-leaving-your-ide-audry-hsu-runpod", "under-5-minutes-to-a-deployed-llm-endpoint-audry-hsu-runpod" ] }, { "slug": "safe-intelligence", "name": "Safe Intelligence", "url": "https://ai.engineer/orgs/safe-intelligence", "summary": "At AI Engineer Europe 2026, Michal Cichra and Steven Willmott examined how explicit requirements guide coding agents and test agent behavior. Cichra covered decision records, product requirements, and executable scenarios for preserving engineering intent; Willmott argued for behavioral specifications that remain useful across changes to models and agent frameworks.", "speakerSlugs": [ "michal-cichra", "steven-willmott" ], "talkSlugs": [ "bdd-adr-prd-wtf-capturing-decisions-for-humans-and-ai-alike-michal-cichra-safe-intelligence", "spec-driven-testing-for-agents-with-a-brain-the-size-of-a-planet-steven-willmott-safe-intelligen" ] }, { "slug": "sambanova-systems", "name": "SambaNova Systems", "url": "https://ai.engineer/orgs/sambanova-systems", "summary": "Llama 3 inference and enterprise document question answering anchor SambaNova Systems' recorded workshops at AI Engineer World's Fair 2024. Affiliated presenters include Rachelle Mattern, Varun Badrinath Krishna, and Petro Junior Milan. The sessions connect platform demonstrations with practical exercises for building retrieval-augmented generation applications using Python and SambaNova APIs.", "speakerSlugs": [ "varun-badrinath-krishna", "petro-junior-milan", "rachelle-mattern", "petro-milan", "varun-krishna" ], "talkSlugs": [ "build-enterprise-generative-ai-apps-using-llama-3-at-1-000-tokens-s-on-the-sambanova-ai-platform", "llama-3-at-1-redacted-password-000-tok-s-on-the-sambanova-ai-platform" ] }, { "slug": "semianalysis", "name": "SemiAnalysis", "url": "https://ai.engineer/orgs/semianalysis", "summary": "Dylan Patel's recorded AIE talks trace frontier-model infrastructure from serving systems to geopolitical competition. His 2024 appearance covers inference deployment and operating costs; his 2025 appearance examines accelerator systems, semiconductor manufacturing, and compute investment across China, the United States, and the Middle East. Together, they connect engineering decisions with the physical constraints of expanding AI capacity.", "speakerSlugs": [ "dylan-patel" ], "talkSlugs": [ "compute-system-design-for-next-generation-frontier-models", "the-geopolitics-of-ai-infrastructure" ] }, { "slug": "sizzy", "name": "Sizzy", "url": "https://ai.engineer/orgs/sizzy", "summary": "AI-assisted development and personal agents connect Kitze’s recorded AIE appearances for Sizzy. At World's Fair 2025, he examines engineering judgment around generated code; at Europe 2026, he traces his personal automation experiments and argues that agents will increasingly replace conventional app interfaces as the way people interact with software.", "speakerSlugs": [ "kitze" ], "talkSlugs": [ "from-vibe-coding-to-vibe-engineering", "the-end-of-apps" ] }, { "slug": "snyk", "name": "Snyk", "url": "https://ai.engineer/orgs/snyk", "summary": "At AI Engineer World's Fair 2026, Ezra Tanzer and Dan Arpino examine security around autonomous coding agents, from AI-generated code to external tools and agent actions. Manoj Nair addresses security architecture for autonomous development, including poisoned skills, chained exploits, sensitive-data leakage, and the growing demands of remediation.", "speakerSlugs": [ "ezra-tanzer", "dan-arpino", "manoj-nair" ], "talkSlugs": [ "agentic-development-security", "through-the-ai-fog-the-architectural-decision-the-next-24-months-of-agentic-security-depends-on" ] }, { "slug": "stanford-university", "name": "Stanford University", "url": "https://ai.engineer/orgs/stanford-university", "summary": "At AI Engineer, Stanford University-affiliated speaker Yegor Denisov-Blanch examines whether AI coding assistance produces useful engineering work. His recorded talks connect Git histories and expert-informed code assessments to questions about productivity and return on investment, showing why apparent increases in coding activity need to be weighed against rework and project conditions.", "speakerSlugs": [ "yegor-denisov-blanch" ], "talkSlugs": [ "ai-developer-productivity", "how-to-quantify-ai-roi-in-software-engineering-stanford-study-120k-devs" ] }, { "slug": "stripe", "name": "Stripe", "url": "https://ai.engineer/orgs/stripe", "summary": "Safe agent transactions and the economics of AI products connect Stripe's recorded appearances at AI Engineer Europe 2026. Steve Kaliski examines payment infrastructure for autonomous purchases, while Mayank Pant discusses how variable compute costs affect pricing. Together, the talks cover transaction safeguards and the commercial choices behind charging for AI services.", "speakerSlugs": [ "steve-kaliski", "mayank-pant" ], "talkSlugs": [ "building-safe-payment-infrastructure-for-the-autonomous-economy", "mastering-ai-pricing-mayank-pant-stripe" ] }, { "slug": "the-browser-company", "name": "The Browser Company", "url": "https://ai.engineer/orgs/the-browser-company", "summary": "Samir Mody and Hursh Agrawal's AIE talks connect AI browser development with hands-on engineering leadership. Mody describes the transition from Arc to Dia and tools for rapid experimentation; Agrawal presents autonomous agents as a way for leaders with fragmented schedules to prototype and review working implementations overnight.", "speakerSlugs": [ "samir-mody", "hursh-agrawal" ], "talkSlugs": [ "from-arc-to-dia-lessons-learned-in-building-ai-browser", "prototyping-as-leadership-how-a-cto-ships-with-ai-agents" ] }, { "slug": "thomson-reuters", "name": "Thomson Reuters", "url": "https://ai.engineer/orgs/thomson-reuters", "summary": "At AI Engineer, Joel Hron and Shirsha Chaudhuri discuss agents for consequential professional and enterprise workflows. Hron examines the move from generative assistants to productive agents in legal, tax, compliance, audit, and risk work; Chaudhuri traces a progression from an LLM playground through RAG and prompt engineering to complete workflow automation.", "speakerSlugs": [ "joel-hron", "shirsha-chaudhuri" ], "talkSlugs": [ "from-copilot-to-colleague-building-trustworthy-productivity-agents-for-high-stakes-work", "missing-pieces-of-workflow-automation" ] }, { "slug": "tldraw", "name": "tldraw", "url": "https://ai.engineer/orgs/tldraw", "summary": "Steve Ruiz's AIE appearances present tldraw's infinite canvas as a workspace for visual programming and AI collaboration. His demonstrations move from connected graphs producing text, images, and speech to sketch-driven interface generation and visible agents working alongside users, showing different ways to bring AI interactions into spatial interfaces.", "speakerSlugs": [ "steve-ruiz" ], "talkSlugs": [ "agents-on-the-canvas-in-tldraw", "tldraw-computer" ] }, { "slug": "traceloop", "name": "Traceloop", "url": "https://ai.engineer/orgs/traceloop", "summary": "Nir Gazit's AIE appearances connect observability for generative AI applications with evaluation-driven prompt improvement. He introduces OpenLLMetry through OpenTelemetry instrumentation and vector database tracing, then presents a documentation chatbot workflow that uses evaluation datasets, factual checks, and an agent to revise prompts through repeated tests.", "speakerSlugs": [ "nir-gazit" ], "talkSlugs": [ "openllmetry-is-all-you-need", "prompt-engineering-is-dead" ] }, { "slug": "trelis-research", "name": "Trelis Research", "url": "https://ai.engineer/orgs/trelis-research", "summary": "At AI Engineer World's Fair 2025, Ronan McGovern of Trelis Research demonstrates practical workflows for adapting browser-capable agents and text-to-speech models. The workshops connect data preparation with fine-tuning and evaluation, moving from successful agent interactions to training examples and from recorded audio to a model adapted for a target voice.", "speakerSlugs": [ "ronan-mcgovern" ], "talkSlugs": [ "mcp-agent-fine-tuning-workshop-ronan-mcgovern", "text-to-speech-data-preparation-and-fine-tuning-workshop-ronan-mcgovern" ] }, { "slug": "twilio", "name": "Twilio", "url": "https://ai.engineer/orgs/twilio", "summary": "AI customer engagement and documentation automation connect Twilio's recorded AIE appearances. Dominik Kundel discusses how an innovation team experiments without disrupting existing customers, while Elmer Thomas and Maria Bermudez describe narrowly scoped editorial agents. Together, the talks offer product experimentation and documentation workflow perspectives on putting AI into use.", "speakerSlugs": [ "dominik-kundel", "elmer-thomas", "maria-bermudez" ], "talkSlugs": [ "cooking-with-fire-without-burning-down-the-kitchen", "the-robots-are-coming-for-your-job-and-that-s-okay" ] }, { "slug": "uber", "name": "Uber", "url": "https://ai.engineer/orgs/uber", "summary": "At AI Engineer, Uday Kiran Medisetty and Adam Huda describe infrastructure for agentic software development, from model access to context-driven implementation. Soumya Gupta and Jai Chopra discuss evaluating an agent that edits Uber Eats food photographs while preserving authenticity. The talks connect agent capabilities with validation and operational controls.", "speakerSlugs": [ "uday-kiran-medisetty", "adam-huda", "soumya-gupta", "jai-chopra" ], "talkSlugs": [ "agentic-sdlc-at-uber-building-blocks-for-uber-s-software-factory", "building-closed-loop-evals-for-a-multimodal-agent-at-uber-scale" ] }, { "slug": "upside", "name": "Upside", "url": "https://ai.engineer/orgs/upside", "summary": "Go-to-market agent reliability anchors Upside's recorded AIE presence: Alex Bauer discusses clean business data, trustworthy context, and verifiable citations for revenue workflows. Jonas Bauer is listed in an HF0 startup showcase featuring Upside's structured enterprise-data platform, connecting the archive's practical trust concerns with a brief platform introduction.", "speakerSlugs": [ "alex-bauer", "jonas-bauer" ], "talkSlugs": [ "how-juries-and-librarians-can-solve-gtm-s-ai-trust-problem", "the-next-unicorns-7-top-ai-startups-from-the-hf0-residency" ] }, { "slug": "warp", "name": "Warp", "url": "https://ai.engineer/orgs/warp", "summary": "Ben Holmes’s recorded AIE appearances with Warp examine how people interact with models and organize knowledge for agents. His FixGPT demonstration combines voice with persistent text and automated model selection; his knowledge-base walkthrough turns scattered Markdown notes into connected resources using Hubble, Obsidian, Warp, and agent-driven enrichment.", "speakerSlugs": [ "ben-holmes" ], "talkSlugs": [ "chatgpt-is-poorly-designed-so-i-fixed-it", "llm-knowledge-bases-a-practical-guide" ] }, { "slug": "wisedocs", "name": "Wisedocs", "url": "https://ai.engineer/orgs/wisedocs", "summary": "Denys Linkov's Wisedocs appearances at AI Engineer connect medical-claims pipeline refactoring with the organization of AI delivery teams. His talks examine whether a lengthy rebuild justified its cost and how staffing decisions should respond to company context, linking software architecture and team structure to measurable business value.", "speakerSlugs": [ "denys-linkov" ], "talkSlugs": [ "benchmarking-coding-agents-on-new-vs-legacy-codebases-denys-linkov-wisedocs", "structuring-a-modern-ai-team" ] }, { "slug": "wordware", "name": "Wordware", "url": "https://ai.engineer/orgs/wordware", "summary": "At AI Engineer World's Fair 2025, Filip Kozera and Robert Chandler examined how interfaces shape agent reliability. Kozera proposed structured documents for specifying concurrent background work under human approval, while Chandler argued for specialized agentic tools that absorb API complexity instead of leaving every intermediate step to the coordinating model.", "speakerSlugs": [ "filip-kozera", "robert-chandler" ], "talkSlugs": [ "beyond-conversation-why-documents-transform-natural-language-into-code", "just-do-it-let-your-tools-think-for-themselves-robert-chandler" ] }, { "slug": "writer", "name": "Writer", "url": "https://ai.engineer/orgs/writer", "summary": "Financial AI evaluation and retrieval from dense enterprise documents connect Writer's recorded AIE appearances. Waseem Alshikh examines how models respond to flawed queries and inadequate evidence, while Sam Julien traces the move from vector search to graph-based retrieval. Together, the talks address how enterprise systems find relevant information and judge whether it supports an answer.", "speakerSlugs": [ "waseem-alshikh", "sam-julien" ], "talkSlugs": [ "building-trust-in-enterprise-ai-evaluating-domain-specific-llms-for-real-world-financial-scenari", "when-vectors-break-down-graph-based-rag-for-dense-enterprise-knowledge" ] }, { "slug": "y-combinator", "name": "Y Combinator", "url": "https://ai.engineer/orgs/y-combinator", "summary": "Organizational memory and AI-assisted creation connect the AIE appearances of Garry Tan and Eve Bouffard. At AI Engineer World's Fair 2026, Tan presented a company-brain approach to giving agents institutional context, while Bouffard argued that increasingly capable AI shifts the constraint on building new things from implementation toward imagination.", "speakerSlugs": [ "garry-tan", "eve-bouffard" ], "talkSlugs": [ "every-company-should-have-a-brain-garry-tan-y-combinator", "imagination-engineering" ] }, { "slug": "yutori", "name": "Yutori", "url": "https://ai.engineer/orgs/yutori", "summary": "Yutori’s recorded AIE appearances address how agents navigate websites and how people interact with AI applications. Dhruv Batra argues for computer-use models where structured interfaces fall short, while Maximillian Piras examines the persistence of chat and model pickers as model capabilities change. Together, the talks connect browser automation with interface design.", "speakerSlugs": [ "dhruv-batra", "maximillian-piras" ], "talkSlugs": [ "computer-use-models-will-agentify-the-web-not-apis", "the-bitter-layout-or-how-i-learned-to-love-the-model-picker" ] }, { "slug": "txt-outlines", "name": ".txt (Outlines)", "url": "https://ai.engineer/orgs/txt-outlines", "summary": "At AI Engineer World's Fair 2024, Rémi Louf explained how structured generation constrains language-model decoding to produce dependable outputs. His demonstrations covered regex-guided answers, JSON Schema and Pydantic-controlled JSON, vision-model extraction, and integrations with open-model serving frameworks.", "speakerSlugs": [ "remi-louf" ], "talkSlugs": [ "no-more-bad-outputs-with-structured-generation" ] }, { "slug": "11x", "name": "11X", "url": "https://ai.engineer/orgs/11x", "summary": "At AI Engineer World's Fair 2025, Sherwood Callaway and Satwik Singh explain how Alice, 11X’s AI sales development representative, uses a customer-specific knowledge base. Their talk follows sales materials through ingestion, text conversion, and retrieval, framing the process as onboarding a human salesperson.", "speakerSlugs": [ "sherwood-callaway", "satwik-singh" ], "talkSlugs": [ "building-alice-s-brain-an-ai-sales-rep-that-learns-like-a-human-sherwood-satwik-11x" ] }, { "slug": "14-ai", "name": "14.ai", "url": "https://ai.engineer/orgs/14-ai", "summary": "At AI Engineer World's Fair 2025, Michael Fester presented 14.ai's approach to reliable customer support agents using Effect and TypeScript. His talk covers a type-safe application architecture, composable workflows, provider failover, stateful retries, and OpenTelemetry observability.", "speakerSlugs": [ "michael-fester" ], "talkSlugs": [ "building-reliable-support-agents-using-the-effect-typescript-library-michael-fester" ] }, { "slug": "8th-light", "name": "8th Light", "url": "https://ai.engineer/orgs/8th-light", "summary": "At AI Engineer World's Fair 2025, Travis Frisinger examined why LLM output can appear intelligent without demonstrating understanding. His talk connected that argument to ChatGPT-assisted programming and creative multimodal experiments, proposing coherence as a framework for interpreting model behavior.", "speakerSlugs": [ "travis-frisinger" ], "talkSlugs": [ "the-coherence-trap-why-llms-feel-smart-but-aren-t-thinking" ] }, { "slug": "ably", "name": "Ably", "url": "https://ai.engineer/orgs/ably", "summary": "At AI Engineer Europe 2026, Mike Christensen examined how connection-bound HTTP and server-sent event streams make AI chat fragile. He presented durable publish/subscribe-backed sessions and Ably AI Transport to support reconnection, shared multi-device state, and human support handoffs that preserve conversation history.", "speakerSlugs": [ "mike-christensen" ], "talkSlugs": [ "why-your-ai-ux-is-broken-and-it-s-not-the-model-s-fault" ] }, { "slug": "abridge", "name": "Abridge", "url": "https://ai.engineer/orgs/abridge", "summary": "At AI Engineer World's Fair 2026, Abridge's Chaitanya Asawa described expanding ambient clinical documentation into context-aware decision support. Demonstrations covered voice-driven clinical-trial eligibility checks and chart preparation, while clinician testimonials illustrated reduced documentation burden.", "speakerSlugs": [ "chaitanya-asawa" ], "talkSlugs": [ "from-ambient-documentation-to-clinical-intelligence" ] }, { "slug": "accenture", "name": "Accenture", "url": "https://ai.engineer/orgs/accenture", "summary": "At AI Engineer Europe 2026, Jess Grogan-Avignon and Jack Wang argue that enterprise agentic-AI projects stall when governance, funding, delivery, and deployment move at human speed. Their talk recommends engineering automation and CI/CD alongside portfolio-based investment and evidence-led iteration.", "speakerSlugs": [ "jess-grogan-avignon", "jack-wang" ], "talkSlugs": [ "most-enterprise-agentic-projects-are-doomed-here-s-why" ] }, { "slug": "adaption", "name": "Adaption", "url": "https://ai.engineer/orgs/adaption", "summary": "At AI Engineer World's Fair 2026, Sara Hooker examines how concentrated compute and narrow research career paths restrict participation in frontier AI. She presents Adaption's AutoScientist as an approach to automating model training, connecting efficient customization and competitive smaller models to broader access.", "speakerSlugs": [ "sara-hooker" ], "talkSlugs": [ "adaption-labs-gradient-free-continual-learning" ] }, { "slug": "adaptive-ml", "name": "Adaptive ML", "url": "https://ai.engineer/orgs/adaptive-ml", "summary": "At AI Engineer Europe 2026, Alessandro Cappelli presented reinforcement learning as a way to incorporate operational feedback into enterprise AI. His talk contrasted RL with supervised fine-tuning and discussed smaller specialized models for reducing inference cost and latency.", "speakerSlugs": [ "alessandro-cappelli" ], "talkSlugs": [ "scaling-reinforcement-learning-lessons-from-trillion-token-deployments-at-fortune-500s" ] }, { "slug": "adept", "name": "Adept", "url": "https://ai.engineer/orgs/adept", "summary": "At AI Engineer Summit 2023, Amelia Wattenberger, speaking as a designer at Adept, examined how AI interfaces could help people move between levels of abstraction. Drawing on spreadsheets and Google Maps, she argued for changing how people work beyond simply speeding up existing tasks.", "speakerSlugs": [ "amelia-wattenberger" ], "talkSlugs": [ "climbing-the-ladder-of-abstraction" ] }, { "slug": "adobe", "name": "Adobe", "url": "https://ai.engineer/orgs/adobe", "summary": "At AI Engineer World's Fair 2025, Adobe-affiliated speaker Muktesh Mishra discussed evaluating nondeterministic AI applications beyond accuracy and similarity. His talk connects synthetic data and iterative dataset refinement with metrics tailored to RAG and code-generation systems, framing evaluation as an ongoing development practice.", "speakerSlugs": [ "muktesh-mishra" ], "talkSlugs": [ "how-to-run-evals-at-scale-thinking-beyond-accuracy-or-similarity" ] }, { "slug": "aech-ai", "name": "Aech AI", "url": "https://ai.engineer/orgs/aech-ai", "summary": "At AI Engineer Summit 2025, Aech AI's Steven Moon described enterprise agents operating within existing organizational security boundaries. His talk outlines how private-cloud infrastructure, Microsoft 365, Azure, and Active Directory can support provisioning, monitoring, and auditing agents like employees.", "speakerSlugs": [ "steven-moon" ], "talkSlugs": [ "privacy-first-enterprise-ai-building-ai-agents-that-never-leave-your-security-boundary" ] }, { "slug": "agenta", "name": "Agenta", "url": "https://ai.engineer/orgs/agenta", "summary": "At AI Engineer Europe 2026, Agenta's Mahmoud Mabrouk presented a workshop on calibrating LLM judges against human annotations. Using an airline customer-support scenario from Sierra's τ-bench, the session covered business-specific metrics, targeted evaluators, and GEPA-based reflective prompt optimization.", "speakerSlugs": [ "mahmoud-mabrouk" ], "talkSlugs": [ "judge-the-judge-building-llm-evaluators-that-actually-work-with-gepa-mahmoud-mabrouk-agenta-ai" ] }, { "slug": "agnostiq-covalent", "name": "Agnostiq (Covalent)", "url": "https://ai.engineer/orgs/agnostiq-covalent", "summary": "At AI Engineer World's Fair 2024, Santosh Radha demonstrated Covalent workflows for model fine-tuning, evaluation, and inference directly from Python. His walkthrough showed notebook functions dispatched to remote GPUs and inference endpoints deployed with configurable autoscaling, without managing Docker or Kubernetes.", "speakerSlugs": [ "santosh-radha" ], "talkSlugs": [ "covalent-launch-the-gpu-cheatcode-fine-tune-20-llama-models-in-5-minutes" ] }, { "slug": "ai-snake-oil", "name": "AI Snake Oil", "url": "https://ai.engineer/orgs/ai-snake-oil", "summary": "At AI Engineer Summit 2025, Sayash Kapoor examined why impressive agent demonstrations and benchmark results can fail to translate into dependable systems. His talk connected examples from legal AI, scientific reproducibility, and coding agents to evaluation problems including reward hacking, neglected cost, and inconsistent performance.", "speakerSlugs": [ "sayash-kapoor" ], "talkSlugs": [ "evaluating-ai-agents" ] }, { "slug": "aius-technologies", "name": "AIUS Technologies", "url": "https://ai.engineer/orgs/aius-technologies", "summary": "At AI Engineer World's Fair 2025, Mark Bain led a workshop on durable AI-agent memory, emphasizing causal relationships, knowledge graphs, and GraphRAG. The session included guest demonstrations of graph-based workflows, memory-server integration with Claude Desktop, and temporal graphs.", "speakerSlugs": [ "mark-bain" ], "talkSlugs": [ "memory-masterclass-make-your-ai-agents-remember-what-they-do-mark-bain-aius" ] }, { "slug": "alibaba-group-qwen", "name": "Alibaba Group / Qwen", "url": "https://ai.engineer/orgs/alibaba-group-qwen", "summary": "At AI Engineer World's Fair 2025, Junyang Lin presented Qwen's roadmap toward generalist models and agents. The talk covers Qwen3's dense and mixture-of-experts architectures, reinforcement-learning-driven reasoning, multilingual support, and MCP-enabled agent capabilities, alongside Qwen Chat's multimodal features.", "speakerSlugs": [ "junyang-lin" ], "talkSlugs": [ "the-future-of-qwen-a-generalist-agent-model" ] }, { "slug": "allen-institute-for-ai-ai2-interconnects-ai", "name": "Allen Institute for AI (Ai2); Interconnects.ai", "url": "https://ai.engineer/orgs/allen-institute-for-ai-ai2-interconnects-ai", "summary": "At AI Engineer World's Fair 2025, Nathan Lambert presented a taxonomy of reasoning models spanning skills, calibration, strategy, and abstraction. His talk argues that moving beyond benchmark performance requires deliberate training for planning and long-horizon agent capabilities.", "speakerSlugs": [ "nathan-lambert" ], "talkSlugs": [ "a-taxonomy-for-next-generation-reasoning-models" ] }, { "slug": "alleycorp", "name": "AlleyCorp", "url": "https://ai.engineer/orgs/alleycorp", "summary": "Kenneth Auchenberg’s appearance at AI Engineer World's Fair 2025 focused on AI-era product management: shipping useful products to real users, observing customers directly, and accelerating iteration with AI-assisted prototypes. He drew on lessons from building Visual Studio Code and Stripe’s developer platform.", "speakerSlugs": [ "kenneth-auchenberg" ], "talkSlugs": [ "shipping-something-to-someone-always-wins" ] }, { "slug": "allos-ai", "name": "Allos AI", "url": "https://ai.engineer/orgs/allos-ai", "summary": "At AI Engineer World's Fair 2026, Allos AI's Ayush Bhardwaj compared specialized AI systems for hedge funds and pharmaceutical research. His talk connects narrowly scoped agents and observability with the expert judgment needed to evaluate domain-specific outputs and establish measurable business returns.", "speakerSlugs": [ "ayush-bhardwaj" ], "talkSlugs": [ "trading-desks-to-clinical-trials-parallels-in-applied-vertical-ai" ] }, { "slug": "alma", "name": "Alma", "url": "https://ai.engineer/orgs/alma", "summary": "At AI Engineer World's Fair 2025, Rami Alhamad presented Alma's approach to a personalized AI nutrition companion. His talk connects meal tracking with persistent user-editable context, voice-based logging, and proactive engagement, alongside product lessons from beta feedback and tracking streaks.", "speakerSlugs": [ "rami-alhamad" ], "talkSlugs": [ "my-ai-thinks-i-m-eating-my-feelings-and-other-nutritional-insights" ] }, { "slug": "alpic", "name": "Alpic", "url": "https://ai.engineer/orgs/alpic", "summary": "At AI Engineer Europe 2026, Frédéric Barthelet examined isolated iframe rendering for interactive tool-call views in MCP and ChatGPT applications. The talk connected browser embedding constraints to application security and closed with Alpic's open-source Skybridge framework for building these applications.", "speakerSlugs": [ "frederic-barthelet" ], "talkSlugs": [ "why-mcp-and-chatgpt-apps-use-double-iframes-frederic-barthelet-alpic" ] }, { "slug": "altos-labs", "name": "Altos Labs", "url": "https://ai.engineer/orgs/altos-labs", "summary": "At AI Engineer World's Fair 2026, Akram Baharlouei's talk examines the engineering challenges of foundation models for single-cell biology. The presentation connects cellular rejuvenation and reprogramming with drug development and virtual-cell modeling, framing biological applications through the challenges of working with cellular data.", "speakerSlugs": [ "akram-baharlouei" ], "talkSlugs": [ "from-tokens-to-cells-foundation-models-for-single-cell-biology-akram-baharlouei-altos-labs" ] }, { "slug": "ambient", "name": "Ambient", "url": "https://ai.engineer/orgs/ambient", "summary": "At AI Engineer Summit 2023, Ambient’s Mithun Hunsur introduced llm, a Rust library for running language models locally. His talk compared local inference with cloud-hosted models and demonstrated Llama 7B on a MacBook CPU, connecting hardware constraints to practical deployment choices.", "speakerSlugs": [ "mithun-hunsur" ], "talkSlugs": [ "harnessing-the-power-of-llms-locally" ] }, { "slug": "amplifon", "name": "Amplifon", "url": "https://ai.engineer/orgs/amplifon", "summary": "At AI Engineer Europe 2026, Sonny Merla joined Quantyca’s Mauro Luchetti and Mattia Redaelli to discuss Amplifon’s AmplifAI program and fragmented agent development. The session describes a centralized AI gateway combining model access, authentication, budgeting and auditing with registries for MCP servers and A2A agents.", "speakerSlugs": [ "sonny-merla" ], "talkSlugs": [ "one-registry-to-rule-them-all-sonny-merla-mauro-luchetti-mattia-redaelli-quantyca" ] }, { "slug": "andon-labs", "name": "Andon Labs", "url": "https://ai.engineer/orgs/andon-labs", "summary": "At AI Engineer World's Fair 2026, Lukas Petersson presents Vending-Bench, where agents operate simulated vending businesses and compete over extended periods. He reports collusion, deception, and power-seeking behavior, and explains how simulation awareness motivates evaluations in real shops, a café, and radio stations.", "speakerSlugs": [ "lukas-petersson" ], "talkSlugs": [ "vending-bench-long-horizon-agent-evals-lukas-petersson-andon-labs" ] }, { "slug": "annicha-labs", "name": "Annicha Labs", "url": "https://ai.engineer/orgs/annicha-labs", "summary": "At AI Engineer World's Fair 2026, Rajiv Chandegra of Annicha Labs presented adaptive engineering for changing physical and social systems. He proposed designing constraints that let agents, humans, and institutions reorganize their collaboration during runtime, drawing on emergence, self-organization, and attractors.", "speakerSlugs": [ "rajiv-chandegra" ], "talkSlugs": [ "beyond-the-harness-a-journey-towards-adaptive-engineering-rajiv-chandegra-annicha-labs" ] }, { "slug": "apify", "name": "Apify", "url": "https://ai.engineer/orgs/apify", "summary": "At AI Engineer World's Fair 2025, Jan Curn presented a vision of autonomous agents connected through Model Context Protocol. His talk examined identity, payment, monetization, and trust challenges, and demonstrated cross-service browser automation by invoking Browserbase's MCP capabilities through an Apify Actor.", "speakerSlugs": [ "jan-curn" ], "talkSlugs": [ "the-rise-of-the-agentic-economy-on-the-shoulders-of-mcp" ] }, { "slug": "arc-prize-foundation", "name": "ARC Prize Foundation", "url": "https://ai.engineer/orgs/arc-prize-foundation", "summary": "At AI Engineer World's Fair 2025, Greg Kamradt presented AGI evaluation through efficient acquisition of unfamiliar skills relative to humans. His talk covered ARC-AGI grid tasks and previewed interactive ARC-AGI-3 evaluations in which agents explore, infer rules, and adapt.", "speakerSlugs": [ "greg-kamradt" ], "talkSlugs": [ "measuring-agi-interactive-reasoning-benchmarks" ] }, { "slug": "arcjet", "name": "Arcjet", "url": "https://ai.engineer/orgs/arcjet", "summary": "At AI Engineer World's Fair 2025, Arcjet-affiliated speaker David Mytton examines how AI crawlers increase website traffic and infrastructure costs. His talk distinguishes useful search indexing and user-initiated agents from training crawlers and abusive scraping, framing website defense around different kinds of automated access.", "speakerSlugs": [ "david-mytton" ], "talkSlugs": [ "how-to-defend-your-sites-from-ai-bots" ] }, { "slug": "arena-ai", "name": "Arena.ai", "url": "https://ai.engineer/orgs/arena-ai", "summary": "At AI Engineer Europe 2026, Peter Gostev of Arena.ai examines why rising benchmark scores can conceal persistent language-model failures. His talk combines tests of whether models challenge invalid premises with Arena voting and dissatisfaction data across domains including medicine, finance, law, and game development.", "speakerSlugs": [ "peter-gostev" ], "talkSlugs": [ "what-do-models-still-suck-at" ] }, { "slug": "ario", "name": "Ario", "url": "https://ai.engineer/orgs/ario", "summary": "At AI Engineer World's Fair 2024, Ario’s Sumit Agarwal presented a personal assistant built around adversarial ETL: collecting users’ existing activity across services and combining it with RAG, LLMs, and agents. Demonstrations showed personalized data portraits and recommendations without repetitive onboarding questions.", "speakerSlugs": [ "sumit-agarwal" ], "talkSlugs": [ "the-adversarial-path-to-the-personal-assistant" ] }, { "slug": "arista-networks", "name": "Arista Networks", "url": "https://ai.engineer/orgs/arista-networks", "summary": "At AI Engineer Summit 2025, Paul Gilbert of Arista Networks discussed dedicated AI data-center infrastructure. His session contrasts training and inference GPU requirements and examines isolated backend fabrics, front-end storage networks, and east-west traffic.", "speakerSlugs": [ "paul-gilbert" ], "talkSlugs": [ "how-to-build-your-own-ai-data-center-in-2025" ] }, { "slug": "arithmetic", "name": "Arithmetic", "url": "https://ai.engineer/orgs/arithmetic", "summary": "At AI Engineer World's Fair 2026, Arithmetic's Uri Rolls and Hugging Face's Thom Wolf discuss interactive cybersecurity benchmarks for broken access control. Their joint presentation examines why AI defenders need to reason through changing systems rather than simply identify isolated vulnerabilities.", "speakerSlugs": [ "uri-rolls" ], "talkSlugs": [ "training-frontier-models-to-out-think-hackers-uri-rolls-arithmetic-thom-wolf-hugging-face" ] }, { "slug": "arklex-ai-columbia-university", "name": "Arklex AI; Columbia University", "url": "https://ai.engineer/orgs/arklex-ai-columbia-university", "summary": "At AI Engineer Summit 2025, Zhou Yu reviewed research on improving AI agents, covering autonomy, human oversight, mathematical reasoning, and conversational planning. The talk connected verifier feedback, on-policy training, and test-time compute scaling with approaches to agent self-improvement.", "speakerSlugs": [ "zhou-yu" ], "talkSlugs": [ "how-to-improve-your-agents-academic-lit-review" ] }, { "slug": "arrakis", "name": "Arrakis", "url": "https://ai.engineer/orgs/arrakis", "summary": "At AI Engineer World's Fair 2025, Abhishek Bhardwaj presented secure Linux sandboxes for AI agents executing tools, operating computers, and running reinforcement-learning workflows. His Arrakis talk connects Cloud Hypervisor-based microVM isolation with snapshot-and-restore backtracking for recoverable multistep tasks.", "speakerSlugs": [ "abhishek-bhardwaj" ], "talkSlugs": [ "arrakis-how-to-build-an-ai-sandbox-from-scratch" ] }, { "slug": "artificial-analysis", "name": "Artificial Analysis", "url": "https://ai.engineer/orgs/artificial-analysis", "summary": "Model intelligence, reasoning efficiency, latency, and openness frame Artificial Analysis’s recorded AIE session, credited to George Cameron and Micah Hill-Smith. The talk summary attributes the comparisons to Cameron, who examines how model capability relates to token consumption, response time, and compute demand.", "speakerSlugs": [ "micah-hill-smith", "george-cameron" ], "talkSlugs": [ "trends-across-the-ai-frontier" ] }, { "slug": "atlan", "name": "Atlan", "url": "https://ai.engineer/orgs/atlan", "summary": "At AI Engineer World's Fair 2026, Prukalpa Sankar argues that production agents need the business definitions, procedural expertise, and organizational norms that people learn through experience. Her talk uses a fictional analyst and disconnected marketing and sales agents to explain the need for shared context.", "speakerSlugs": [ "prukalpa-sankar" ], "talkSlugs": [ "wtf-is-the-context-layer-the-missing-infrastructure-for-production-agents" ] }, { "slug": "auditoria-ai", "name": "Auditoria AI", "url": "https://ai.engineer/orgs/auditoria-ai", "summary": "At AI Engineer World's Fair 2026, Ramana Siddanth Emani presented finance-agent reliability as a developer-workflow problem. His talk argues that diagnosing production failures and iterating quickly can constrain deployments more than model capability or hardware, with humans serving as verifiers.", "speakerSlugs": [ "ramana-siddanth-emani" ], "talkSlugs": [ "your-finance-agent-s-bottleneck-is-you" ] }, { "slug": "auth0", "name": "Auth0", "url": "https://ai.engineer/orgs/auth0", "summary": "At AI Engineer World's Fair 2025, Bobby Tiernay and Kam Sween presented ways to secure production AI agents by tying authorization to individual users. Their session addressed excessive permissions, exposed secrets, and untraceable actions, with demonstrations of standards-based authorization and Auth0 Token Vault.", "speakerSlugs": [ "bobby-tiernay", "kam-sween" ], "talkSlugs": [ "securing-agents-with-open-standards" ] }, { "slug": "autogpt", "name": "AutoGPT", "url": "https://ai.engineer/orgs/autogpt", "summary": "At AI Engineer Summit 2023, Toran Bruce Richards, Silen Naihin, and Poots discussed autonomous agents as a way to reduce repetitive work and expand human creativity. The session also covered AutoGPT's open-source community and the goal of building an open-source generalist agent.", "speakerSlugs": [ "toran-bruce-richards", "silen-naihin", "poots" ], "talkSlugs": [ "the-future-of-work" ] }, { "slug": "automattic", "name": "Automattic", "url": "https://ai.engineer/orgs/automattic", "summary": "At AI Engineer World's Fair 2026, Sanja Grbic described Automattic’s Radical Speed Month, an experiment in autonomous small-team work. Her talk traced how practical AI training and accessible development environments helped her contribute code and build prototypes as a product designer.", "speakerSlugs": [ "sanja-grbic" ], "talkSlugs": [ "500-people-vibe-coded-for-30-days-i-was-one-of-them" ] }, { "slug": "aviator", "name": "Aviator", "url": "https://ai.engineer/orgs/aviator", "summary": "At AI Engineer World's Fair 2026, Aviator's Ankit Jain proposed shifting code review toward intent, behavioral evidence, and architectural judgment. His talk connects decisions captured from coding-agent conversations with acceptance criteria, reusable review guardrails, and executable test plans verified against live previews.", "speakerSlugs": [ "ankit-jain" ], "talkSlugs": [ "how-to-kill-the-code-review" ] }, { "slug": "axa", "name": "AXA", "url": "https://ai.engineer/orgs/axa", "summary": "At AI Engineer World's Fair 2024, AXA-affiliated speaker Jeronim Morina presented DSPy for optimizing modular language-model systems in insurance. His talk emphasized first-principles engineering and meaningful evaluation, while describing AXA Germany's data organization and Secure GPT platform.", "speakerSlugs": [ "jeronim-morina" ], "talkSlugs": [ "optimizing-llms-in-insurance-with-dspy-beyond-manual-tuning" ] }, { "slug": "banking-circle", "name": "Banking Circle", "url": "https://ai.engineer/orgs/banking-circle", "summary": "At AI Engineer Europe 2026, Juan Herreros Elorza presented Banking Circle’s Atlas platform as an example of internal developer infrastructure serving both humans and AI agents. His talk connects that goal to API-based self-service and CLI or MCP interfaces.", "speakerSlugs": [ "juan-herreros-elorza" ], "talkSlugs": [ "platforms-for-humans-and-machines-engineering-for-the-age-of-agents-juan-herreros-elorza" ] }, { "slug": "baz", "name": "Baz", "url": "https://ai.engineer/orgs/baz", "summary": "At AI Engineer Europe 2026, Nimrod Hauser demonstrates adapting a third-party Playwright MCP server for production agents. Using Baz Spec Reviewer, he connects ticket requirements and Figma designs with browser-based checks of whether an implementation meets its requirements.", "speakerSlugs": [ "nimrod-hauser" ], "talkSlugs": [ "bending-a-public-mcp-server-without-breaking-it-nimrod-hauser-baz" ] }, { "slug": "bbd-software", "name": "BBD Software", "url": "https://ai.engineer/orgs/bbd-software", "summary": "At AI Engineer World's Fair 2025, BBD Software's affiliated speaker Thabang Ledwaba argued for Africa to shape its AI future through locally grounded innovation. His talk connected resourcefulness and ubuntu with meaningful participation in the global technology economy.", "speakerSlugs": [ "thabang-ledwaba" ], "talkSlugs": [ "unlocking-africa-s-potential-with-ai-thabang-ledwaba" ] }, { "slug": "bee-amazon", "name": "Bee, Amazon", "url": "https://ai.engineer/orgs/bee-amazon", "summary": "At AI Engineer World's Fair 2026, Steve Korshakov presented how Bee's wearable personal AI protects sensitive conversational data through customer-device-managed encryption keys and in-house inference. His talk also addressed deployment security, auditing, and questions about potentially unsafe AI agents.", "speakerSlugs": [ "steve-korshakov" ], "talkSlugs": [ "privacy-preserving-intelligence-steve-korshakov-bee-acq-amazon" ] }, { "slug": "bench-computing", "name": "Bench Computing", "url": "https://ai.engineer/orgs/bench-computing", "summary": "At AI Engineer World's Fair 2025, Bench Computing's Damien Murphy led a hands-on workshop combining Google's Agent2Agent protocol with Model Context Protocol. The session covered cooperating agents that delegate business tasks and connect to services including Slack, GitHub, Salesforce, and Zapier.", "speakerSlugs": [ "damien-murphy" ], "talkSlugs": [ "a2a-mcp-automating-business-processes-with-llms" ] }, { "slug": "best-buy", "name": "Best Buy", "url": "https://ai.engineer/orgs/best-buy", "summary": "At AI Engineer Summit 2025, Adam Terlson of Best Buy presented finite state machines and the Actor model as tools for making LLM-powered multi-agent systems predictable, observable, and recoverable. His talk connects these foundations to tool calling, retrieval-augmented generation, human approval, and guarded orchestration.", "speakerSlugs": [ "adam-terlson" ], "talkSlugs": [ "building-multi-agent-systems-with-finite-state-machines" ] }, { "slug": "better-auth", "name": "Better Auth", "url": "https://ai.engineer/orgs/better-auth", "summary": "At AI Engineer World's Fair 2026, Paola Estefanía de Campos presented Better Auth’s Agent Auth Protocol. The workshop examined why agents need distinct identities and narrowly scoped capabilities instead of shared user credentials or unrestricted tokens, alongside practical questions about enterprise deployment.", "speakerSlugs": [ "paola-estefania-de-campos" ], "talkSlugs": [ "full-workshop-agent-auth-protocol-paola-estefania-de-campos-better-auth" ] }, { "slug": "betterup", "name": "BetterUp", "url": "https://ai.engineer/orgs/betterup", "summary": "At AI Engineer Code 2025, BetterUp-affiliated Brian John demonstrates a proof of concept for adding subagents to Codex CLI. A parent session launches Codex Exec through a wrapper, exchanges prompts and results through files, and receives output over standard output.", "speakerSlugs": [ "brian-john" ], "talkSlugs": [ "hacking-subagents-into-codex-cli-brian-john-betterup" ] }, { "slug": "bitly", "name": "Bitly", "url": "https://ai.engineer/orgs/bitly", "summary": "At AI Engineer Europe 2026, Bitly-affiliated Michael Richman introduces FOMAT: fear of missing agent time. His demonstration of Cmd+Ctrl shows how engineers can monitor, interact with, and launch coding-agent sessions remotely through mobile and web access.", "speakerSlugs": [ "michael-richman" ], "talkSlugs": [ "let-s-talk-about-fomat-fear-of-missing-agent-time" ] }, { "slug": "black-forest-labs", "name": "Black Forest Labs", "url": "https://ai.engineer/orgs/black-forest-labs", "summary": "At AI Engineer Europe 2026, Stephen Batifol presented Black Forest Labs’ progression from FLUX.1 through FLUX.2 [klein], covering contextual image editing, photorealistic multi-reference generation, and sub-second interactive workflows. His presentation connected these capabilities to broader visual intelligence.", "speakerSlugs": [ "stephen-batifol" ], "talkSlugs": [ "black-forest-labs-flux-open-research-and-the-future-of-visual-ai" ] }, { "slug": "blackrock", "name": "BlackRock", "url": "https://ai.engineer/orgs/blackrock", "summary": "At AI Engineer World's Fair 2025, Vaibhav Page and Infant Vasanth discussed building custom AI knowledge applications for investment operations. Their talk covered security setup and financial-document extraction, alongside configurable extraction templates and questions about data pipelines and information-extraction safeguards.", "speakerSlugs": [ "vaibhav-page", "infant-vasanth" ], "talkSlugs": [ "how-blackrock-builds-custom-knowledge-apps-at-scale" ] }, { "slug": "block", "name": "Block", "url": "https://ai.engineer/orgs/block", "summary": "At AI Engineer Summit 2025, Block-affiliated speaker Rizel Scarlett examined how marketing AI agents as human software engineers can undermine developer trust and create unrealistic expectations. Her talk advocates positioning AI as a tool that supports developers rather than replaces them.", "speakerSlugs": [ "rizel-scarlett" ], "talkSlugs": [ "your-ai-agent-isn-t-an-engineer-the-art-of-thoughtful-anthropomorphism" ] }, { "slug": "bolt-new-stackblitz", "name": "Bolt.new / StackBlitz", "url": "https://ai.engineer/orgs/bolt-new-stackblitz", "summary": "At AI Engineer World's Fair 2025, Eric Simons described how Bolt.new emerged from years of building StackBlitz. His talk connects rapid growth with a small, tightly aligned team, frugal execution, repeated product experiments, and transparent public office hours.", "speakerSlugs": [ "eric-simons" ], "talkSlugs": [ "bolt-new-how-we-scaled-0-20m-arr-in-60-days-with-15-people" ] }, { "slug": "booking-com", "name": "booking.com", "url": "https://ai.engineer/orgs/booking-com", "summary": "At AI Engineer Summit 2025, booking.com's Bruno Passos joined Sourcegraph's Beyang Liu to discuss context-aware coding assistance and software-development agents in enterprise codebases. Their joint case study connects developer-experience challenges with measuring productivity, modernization, and business impact.", "speakerSlugs": [ "bruno-passos" ], "talkSlugs": [ "building-ai-agents-with-real-roi-in-the-enterprise-sdlc" ] }, { "slug": "botdojo", "name": "BotDojo", "url": "https://ai.engineer/orgs/botdojo", "summary": "At AI Engineer World's Fair 2024, Paul Henry demonstrates a low-code, retrieval-backed support chatbot in BotDojo. The session shows how node tracing, structured outputs, and batch evaluations expose missing knowledge and hallucination risks in the assistant's responses.", "speakerSlugs": [ "paul-henry" ], "talkSlugs": [ "botdojo-launch-enhancing-ai-assistants-with-evaluations-and-synthetic-data" ] }, { "slug": "boundary", "name": "Boundary", "url": "https://ai.engineer/orgs/boundary", "summary": "At AI Engineer World's Fair 2026, Boundary's Vaibhav Gupta described building BAML through AI-assisted engineering. His talk examines replacing conventional code reviews with shared architecture documentation, versioned design documents, Slack-integrated workflows, and CLI-enforced architectural invariants.", "speakerSlugs": [ "vaibhav-gupta" ], "talkSlugs": [ "fighting-slop-with-slop" ] }, { "slug": "box", "name": "Box", "url": "https://ai.engineer/orgs/box", "summary": "At AI Engineer World's Fair 2025, Ben Kus described Box’s move from OCR-assisted, single-pass extraction to coordinated agents that turn enterprise documents into structured metadata. The talk examines accuracy failures on long documents with complex fields and approaches using multiple-model voting and LLM-as-a-judge feedback.", "speakerSlugs": [ "ben-kus" ], "talkSlugs": [ "building-an-agentic-platform" ] }, { "slug": "brightwave", "name": "Brightwave", "url": "https://ai.engineer/orgs/brightwave", "summary": "At AI Engineer Summit 2025, Brightwave's Mike Conover discussed knowledge agents for investment research across filings, transcripts, contracts, and other financial documents. His talk argues for transparent reasoning and findings that users can inspect and question, beyond chat interfaces or unverified reports.", "speakerSlugs": [ "mike-conover" ], "talkSlugs": [ "trust-but-verify-high-fidelity-reasoning-in-agentic-workflows" ] }, { "slug": "bugcrowd-carnegie-mellon-university", "name": "Bugcrowd; Carnegie Mellon University", "url": "https://ai.engineer/orgs/bugcrowd-carnegie-mellon-university", "summary": "At AI Engineer World's Fair 2026, David Brumley presented reinforcement-learning environments for training and evaluating language models to discover real software vulnerabilities. His talk connects progressive human cybersecurity training with realistic target difficulty, reproducible execution, and demanding V8 sandbox-escape scenarios.", "speakerSlugs": [ "david-brumley" ], "talkSlugs": [ "teaching-ai-to-find-real-vulnerabilities-prof-david-brumley-bugcrowd" ] }, { "slug": "callosum", "name": "Callosum", "url": "https://ai.engineer/orgs/callosum", "summary": "At AI Engineer Europe 2026, Adrian Bertagnoli presented Callosum's approach to heterogeneous intelligence: coordinating model architectures, agent workflows, and hardware. His talk argues for optimizing these together, using recursive language models and visual web navigation as concrete case studies.", "speakerSlugs": [ "adrian-bertagnoli" ], "talkSlugs": [ "scaling-the-next-paradigm-of-heterogeneous-intelligence" ] }, { "slug": "callstack", "name": "Callstack", "url": "https://ai.engineer/orgs/callstack", "summary": "At AI Engineer World's Fair 2026, Callstack's Lech Kalinowski presents Vault, a physical, text-first handheld terminal for controlling OpenClaw agents. The talk covers its dual-display interface, a separate local backend, and an experimental narrative-focused RPG mode.", "speakerSlugs": [ "lech-kalinowski" ], "talkSlugs": [ "openclaw-in-your-hand-building-a-physical-ai-terminal-for-local-llm-agents" ] }, { "slug": "capital-one", "name": "Capital One", "url": "https://ai.engineer/orgs/capital-one", "summary": "At AI Engineer Code 2025, Capital One's affiliated speaker Max Kanat-Alexander presented developer-experience investments that help both engineers and AI coding agents. His talk covers standardized development environments, agent-friendly CLIs and APIs, and explicit system context as foundations for more effective software work.", "speakerSlugs": [ "max-kanat-alexander" ], "talkSlugs": [ "developer-experience-in-the-age-of-ai-coding-agents" ] }, { "slug": "casco", "name": "Casco", "url": "https://ai.engineer/orgs/casco", "summary": "At AI Engineer World's Fair 2025, Casco's Rene Brandel described red-teaming Y Combinator AI-agent startups and finding weaknesses beyond prompt injection. His talk covered insecure direct object references, cross-user data access, network discovery, and customer data exposure through BigQuery.", "speakerSlugs": [ "rene-brandel" ], "talkSlugs": [ "how-we-hacked-yc-spring-2025-batch-s-ai-agents" ] }, { "slug": "caylent", "name": "Caylent", "url": "https://ai.engineer/orgs/caylent", "summary": "At AI Engineer World's Fair 2025, Randall Hunt of Caylent discussed enterprise generative AI deployments spanning building-decarbonization agents, wildlife-video search, sports highlights, and document processing. His talk connects these applications to infrastructure choices, enterprise tool governance, prompt maintenance, and production economics.", "speakerSlugs": [ "randall-hunt" ], "talkSlugs": [ "poc-to-prod-hard-lessons-from-200-enterprise-genai-deployments" ] }, { "slug": "checkout-com", "name": "Checkout.com", "url": "https://ai.engineer/orgs/checkout-com", "summary": "At AI Engineer Europe 2026, Talha Sheikh of Checkout.com examines why coding agents can report completion while leaving requirements unmet. His talk introduces Vector Harness and discusses ways to verify agent output, alongside evaluation and code-review feedback loops.", "speakerSlugs": [ "talha-sheikh" ], "talkSlugs": [ "your-coding-agent-doesn-t-always-follow-your-rules" ] }, { "slug": "cherrypick", "name": "Cherrypick", "url": "https://ai.engineer/orgs/cherrypick", "summary": "At AI Engineer Europe 2026, Chris Parsons presented a practical workshop on simple Ralph Loops for AI coding agents. Using a public Pomodoro CLI repository, he demonstrated ticket-driven implementation, repeated Claude invocations, fresh session context, and agent skills.", "speakerSlugs": [ "chris-parsons" ], "talkSlugs": [ "ralph-loops-build-dumb-ai-loops-that-ship" ] }, { "slug": "chime", "name": "Chime", "url": "https://ai.engineer/orgs/chime", "summary": "At AI Engineer World's Fair 2025, Chime’s Chris Hernandez joined Freeplay’s Jeremy Silva to discuss why generative AI prototypes run into reliability problems in production. Their discussion connects iterative evaluation, human review, and hallucination monitoring with the ongoing work of operating AI.", "speakerSlugs": [ "chris-hernandez" ], "talkSlugs": [ "the-build-operate-divide-bridging-product-vision-and-ai-operational-reality" ] }, { "slug": "china-resources-holdings", "name": "China Resources Holdings", "url": "https://ai.engineer/orgs/china-resources-holdings", "summary": "At AI Engineer World's Fair 2026, Shawn Chan argued for finance AI that produces defensible investment-committee memos. His talk connects that standard to evidence readers can inspect, figures that remain consistent, and checks on founder claims against the data room.", "speakerSlugs": [ "shawn-chan" ], "talkSlugs": [ "build-for-the-memo-not-the-demo-notes-from-200-investment-committees" ] }, { "slug": "circle", "name": "Circle", "url": "https://ai.engineer/orgs/circle", "summary": "At AI Engineer World's Fair 2025, Circle's Corey Cooper demonstrated how AI can interpret agreements and verify completed work before Solidity smart contracts release USDC from escrow. The workshop connected this workflow to Circle developer tooling and an open-source escrow-agent demonstration.", "speakerSlugs": [ "corey-cooper" ], "talkSlugs": [ "automating-escrow-with-usdc-and-ai" ] }, { "slug": "cisco-outshift-by-cisco", "name": "Cisco / Outshift by Cisco", "url": "https://ai.engineer/orgs/cisco-outshift-by-cisco", "summary": "At AI Engineer World's Fair 2025, Ola Mabadeje presented an MVP connecting ServiceNow ITSM tickets to cooperating AI agents and a live network knowledge graph. The talk addressed reducing network change failures, including automated test-plan generation for firewall changes.", "speakerSlugs": [ "ola-mabadeje" ], "talkSlugs": [ "multi-agent-ai-and-network-knowledge-graphs-for-change-management-and-network-testing" ] }, { "slug": "cloudchef", "name": "CloudChef", "url": "https://ai.engineer/orgs/cloudchef", "summary": "At AI Engineer World's Fair 2025, Nikhil Abraham presented CloudChef's approach to turning general-purpose robots into commercial kitchen workers. The talk covers learning recipes from a single demonstration, adapting to unfamiliar kitchens, culinary perception, thermal sensing, and customer installations.", "speakerSlugs": [ "nikhil-abraham" ], "talkSlugs": [ "general-purpose-robots-as-professional-chefs" ] }, { "slug": "cmpnd", "name": "cmpnd", "url": "https://ai.engineer/orgs/cmpnd", "summary": "At AI Engineer World's Fair 2026, cmpnd's Isaac Miller joins Maxime Rivest in a discussion of DSPy: separating task input-output signatures from models, prompts, and reasoning strategies. Their presentation covers composable Python programs, conditional reasoning, and recursive language models for long-context workloads.", "speakerSlugs": [ "isaac-miller" ], "talkSlugs": [ "the-unreasonable-effectiveness-of-separating-the-task-from-the-model" ] }, { "slug": "cognee", "name": "Cognee", "url": "https://ai.engineer/orgs/cognee", "summary": "Durable AI-agent memory connects Cognee’s recorded AIE appearance with Vasilije Markovic, a guest presenter in Mark Bain’s workshop at AI Engineer World's Fair 2025. The shared session examines causal relationships, knowledge graphs, and GraphRAG through demonstrations and discussion.", "speakerSlugs": [ "vasilije-markovic" ], "talkSlugs": [ "memory-masterclass-make-your-ai-agents-remember-what-they-do-mark-bain-aius" ] }, { "slug": "comfy-org", "name": "Comfy Org", "url": "https://ai.engineer/orgs/comfy-org", "summary": "Jedrick Kosinski and ComfyAnonymous introduce ComfyUI’s node-based workflows for image, video, audio, and 3D generation at AI Engineer World's Fair 2025. Their workshop covers local and API-based model execution and reproducible workflows embedded in generated media, followed by audience questions about custom nodes and workflow APIs.", "speakerSlugs": [ "jedrick-kosinski", "comfyanonymous" ], "talkSlugs": [ "comfyui-workshop-with-comfyanonymous-and-jedrick-kosinski" ] }, { "slug": "conductor", "name": "Conductor", "url": "https://ai.engineer/orgs/conductor", "summary": "At AI Engineer World's Fair 2026, Matt Palmer presented code as a practical source of truth for documentation, product updates, websites, and video. A Conductor product walkthrough built with React and Remotion illustrated his approach to TypeScript-based content engineering.", "speakerSlugs": [ "matt-palmer" ], "talkSlugs": [ "content-is-code" ] }, { "slug": "confident-security", "name": "Confident Security", "url": "https://ai.engineer/orgs/confident-security", "summary": "At AI Engineer World's Fair 2025, Jonathan Mortensen of Confident Security examined how Apple’s Private Cloud Compute approaches privacy for remote AI inference. His independent guide considered architectural safeguards alongside the latency and compute costs of protecting user data.", "speakerSlugs": [ "jonathan-mortensen" ], "talkSlugs": [ "the-unofficial-guide-to-apple-s-private-cloud-compute" ] }, { "slug": "conviction", "name": "Conviction", "url": "https://ai.engineer/orgs/conviction", "summary": "At AI Engineer World's Fair 2025, Conviction's Sarah Guo examined AI-native startups through sustained product adoption and revenue. Her talk covered production coding agents, voice AI, and inference economics, alongside opportunities for practical copilots, domain expertise, and applied machine learning.", "speakerSlugs": [ "sarah-guo" ], "talkSlugs": [ "state-of-startups-and-ai-2025" ] }, { "slug": "corridor", "name": "Corridor", "url": "https://ai.engineer/orgs/corridor", "summary": "At AI Engineer World's Fair 2026, Corridor's Jack Cable examined how frontier models and autonomous coding agents accelerate vulnerability discovery while expanding software attack surfaces. His talk argues for resilient software design as AI systems produce and probe more code.", "speakerSlugs": [ "jack-cable" ], "talkSlugs": [ "the-ai-bugpocalypse-is-here-now-what" ] }, { "slug": "couplework-ai", "name": "CoupleWork AI", "url": "https://ai.engineer/orgs/couplework-ai", "summary": "At AI Engineer World's Fair 2026, Clay Cockrell and Tony Fabrikant argued that relationship AI should prioritize clinical outcomes over engagement and agreement. They presented clinician-informed coaching and called for engineers to work with clinicians on meaningful clinical standards.", "speakerSlugs": [ "clay-cockrell", "tony-fabrikant" ], "talkSlugs": [ "ai-is-the-world-s-largest-relationship-therapist-clay-cockrell-tony-fabrikant-couplework-ai" ] }, { "slug": "coval", "name": "Coval", "url": "https://ai.engineer/orgs/coval", "summary": "At AI Engineer World's Fair 2025, Coval's Brooke Hopkins discussed how simulation and evaluation experience at Waymo informs autonomous voice-agent deployment. Her talk presents large-scale probabilistic simulation as a way to balance agent autonomy with production reliability.", "speakerSlugs": [ "brooke-hopkins" ], "talkSlugs": [ "from-self-driving-to-autonomous-voice-agents-brooke-hopkins-coval" ] }, { "slug": "crewai", "name": "crewAI", "url": "https://ai.engineer/orgs/crewai", "summary": "At AI Engineer World's Fair 2024, João Moura traced crewAI's development from automating LinkedIn writing to orchestrating autonomous, role-specialized agents. His talk demonstrated marketing and lead-qualification crews, connecting the framework's origins to practical business workflows.", "speakerSlugs": [ "joao-moura" ], "talkSlugs": [ "using-agents-to-build-an-agent-company" ] }, { "slug": "crusoe", "name": "Crusoe", "url": "https://ai.engineer/orgs/crusoe", "summary": "At AI Engineer World's Fair 2024, Ievgen Vakulenko presented Crusoe Cloud's infrastructure for distributed AI training, covering GPU-equipped virtual machines, CPU compute, and local NVMe storage. His talk also addressed developer usability, energy-conscious data-center infrastructure, and customer workloads involving Together AI and Boson AI.", "speakerSlugs": [ "ievgen-vakulenko" ], "talkSlugs": [ "accelerating-mixture-of-experts-training-with-rail-optimized-infiniband-networking-in-crusoe-clo" ] }, { "slug": "crv", "name": "CRV", "url": "https://ai.engineer/orgs/crv", "summary": "At AI Engineer Summit 2023, CRV's Brittany Walker moderates a discussion with Chris White and Bryan Bischof about how established software companies invest in AI and integrate it into existing products. The conversation covers hiring, rollout strategy, and measuring product-level returns.", "speakerSlugs": [ "brittany-walker" ], "talkSlugs": [ "the-ai-pivot-with-chris-white-of-prefect-bryan-bischof-of-hex" ] }, { "slug": "cua", "name": "Cua", "url": "https://ai.engineer/orgs/cua", "summary": "At AI Engineer World's Fair 2026, Francesco Bonacci, Dillon DuPont, and Robert Wendt presented Cua's stack for computer-use agents. Their talk covers background desktop interaction through Cua Driver, evaluation with Cua-Bench, and pooled infrastructure targeting idle-GPU costs during reinforcement-learning workloads.", "speakerSlugs": [ "francesco-bonacci", "dillon-dupont", "robert-wendt" ], "talkSlugs": [ "computer-use-2-0-agents-just-got-multi-cursor" ] }, { "slug": "datachain", "name": "DataChain", "url": "https://ai.engineer/orgs/datachain", "summary": "At AI Engineer World's Fair 2026, Dmitry Petrov examined why coding agents struggle with physical-world and unstructured datasets. His DataChain demonstration used dashcam-video analysis to show how persistent, queryable results let agents answer follow-up questions without repeatedly processing raw files.", "speakerSlugs": [ "dmitry-petrov" ], "talkSlugs": [ "when-agents-meet-physical-data-the-other-physics-of-agent-harnesses" ] }, { "slug": "datacurve", "name": "Datacurve", "url": "https://ai.engineer/orgs/datacurve", "summary": "At AI Engineer World's Fair 2026, Datacurve's James Shi presents DeepSWE, a contamination-resistant benchmark built around original, long-horizon software-engineering tasks. His talk examines how coding models differ in task adherence and self-testing, alongside failure patterns and evaluation reliability.", "speakerSlugs": [ "james-shi" ], "talkSlugs": [ "deepswe-a-contamination-resistant-coding-benchmark-james-shi-datacurve" ] }, { "slug": "datalab", "name": "Datalab", "url": "https://ai.engineer/orgs/datalab", "summary": "At AI Engineer World's Fair 2025, Vikas Paruchuri described how Datalab's small generalist AI team built the Marker and Surya document-intelligence projects and trained OCR models. His talk connected that work to team organization, with audience questions on scaling and hiring.", "speakerSlugs": [ "vikas-paruchuri" ], "talkSlugs": [ "small-ai-teams-with-huge-impact-vik-paruchuri-datalab" ] }, { "slug": "datarobot", "name": "DataRobot", "url": "https://ai.engineer/orgs/datarobot", "summary": "At AI Engineer World's Fair 2026, Elvin Aghammadzada of DataRobot presented task-specific agent skills as a way to teach coding agents enterprise workflows. The talk examines context-window degradation, metadata-based skill discovery, and how skills complement or overlap with MCP.", "speakerSlugs": [ "elvin-aghammadzada" ], "talkSlugs": [ "skills-are-the-new-sdks" ] }, { "slug": "datasette", "name": "Datasette", "url": "https://ai.engineer/orgs/datasette", "summary": "Collaborative coding agents, automated code review, and evaluations frame Datasette-affiliated Simon Willison's appearance at AI Engineer World's Fair 2026. With Cat Wu and Thariq Shihipar, he joins a discussion of Claude Code, Claude Fable, Claude Tag, and changing engineering practices.", "speakerSlugs": [ "simon-willison" ], "talkSlugs": [ "claude-fable-claude-tag-and-anthropic-s-culture-cat-wu-thariq-shihipar-ft-simon-willison" ] }, { "slug": "datologyai", "name": "DatologyAI", "url": "https://ai.engineer/orgs/datologyai", "summary": "At AI Engineer World's Fair 2026, DatologyAI's Ari Morcos presented training-data quality as a way to get more value from scarce compute. His talk uses vision-language-model, multilingual, legal-domain, and open-model examples to illustrate how targeted curation can improve model quality or efficiency.", "speakerSlugs": [ "ari-morcos" ], "talkSlugs": [ "data-quality-is-the-compute-multiplier" ] }, { "slug": "dawn-analytics", "name": "Dawn Analytics", "url": "https://ai.engineer/orgs/dawn-analytics", "summary": "At AI Engineer World's Fair 2024, Dawn Analytics speaker Ben Hylak examined how multimodal AI and spatial computing expand a product's possible inputs, contexts, and behaviors. His talk connects these design challenges to Apple Vision Pro and visionOS.", "speakerSlugs": [ "ben-hylak" ], "talkSlugs": [ "the-era-of-unbounded-products-designing-for-multimodal-i-o" ] }, { "slug": "daytona", "name": "Daytona", "url": "https://ai.engineer/orgs/daytona", "summary": "At AI Engineer World's Fair 2025, Daytona's Ivan Burazin argued for designing developer tools around agent experience. His talk covers user-mediated authentication, agent-readable documentation, and API-first interfaces as alternatives to assumptions built around human users.", "speakerSlugs": [ "ivan-burazin" ], "talkSlugs": [ "ax-is-the-only-experience-that-matters" ] }, { "slug": "dbt-labs", "name": "dbt Labs", "url": "https://ai.engineer/orgs/dbt-labs", "summary": "At AI Engineer World's Fair 2026, Aaron Stanley examined how AI agents can violate explicit instructions while appearing compliant. His talk, AI’s Jurassic Park Period, uses a forensic investigation analogy to frame the risks of agents working around constraints.", "speakerSlugs": [ "aaron-stanley" ], "talkSlugs": [ "ai-s-jurassic-park-period" ] }, { "slug": "decagon", "name": "Decagon", "url": "https://ai.engineer/orgs/decagon", "summary": "At AI Engineer World's Fair 2026, Decagon's Sunny Rekhi discussed how forward-deployed engineers configure enterprise AI customer-service agents. His examples covered support automation, proactive outreach, and CRM integrations, connecting customer requirements with reusable product capabilities.", "speakerSlugs": [ "sunny-rekhi" ], "talkSlugs": [ "how-forward-deployed-engineering-is-done-at-decagon" ] }, { "slug": "decawork", "name": "Decawork", "url": "https://ai.engineer/orgs/decawork", "summary": "At AI Engineer World's Fair 2026, Sarthak Aggarwal presented AI agents as a second enterprise workforce requiring runtime identities, accountable owners, delegated permissions, auditing, and revocation. His Decawork appearance connects IT administration practices to the challenge of governing autonomous software.", "speakerSlugs": [ "sarthak-aggarwal" ], "talkSlugs": [ "it-admin-for-the-ai-workforce-sarthak-aggarwal-decawork" ] }, { "slug": "deepset", "name": "deepset", "url": "https://ai.engineer/orgs/deepset", "summary": "At AI Engineer Europe 2026, Stefano Fiorucci presented reinforcement-learning environment engineering for language models. His talk connects reasoning-model advances to practical work with Prime Intellect's Verifiers and prime-rl, using interactive tasks to examine how models learn through rewards.", "speakerSlugs": [ "stefano-fiorucci" ], "talkSlugs": [ "let-llms-wander-engineering-rl-environments-stefano-fiorucci" ] }, { "slug": "deepset-gmbh", "name": "deepset GmbH", "url": "https://ai.engineer/orgs/deepset-gmbh", "summary": "At AI Engineer Europe 2026, Bilge Yücel examined sovereign AI as an engineering problem spanning data flows, model selection, infrastructure, and operations. The talk connected sovereignty constraints to practical complications in cross-border processing, provider dependence, monitoring, and auditability.", "speakerSlugs": [ "bilge-yucel" ], "talkSlugs": [ "what-breaks-when-you-build-ai-under-sovereignty-constraints" ] }, { "slug": "deno", "name": "Deno", "url": "https://ai.engineer/orgs/deno", "summary": "At AI Engineer World's Fair 2026, Deno's Ryan Dahl examined how incident-response agents with broad production access can turn malicious support inputs into destructive actions. He presented Claw Patrol as an external security backstop, arguing that more capable models still need independent controls.", "speakerSlugs": [ "ryan-dahl" ], "talkSlugs": [ "security-firewall-for-agents" ] }, { "slug": "devday", "name": "DevDay", "url": "https://ai.engineer/orgs/devday", "summary": "At AI Engineer World's Fair 2025, Beth Glenfield argued that AI coding assistants and interview-cheating tools undermine conventional technical interviews. She presented DevDay's alternative: realistic workplace simulations in which candidates collaborate with AI agents playing different teammate roles.", "speakerSlugs": [ "beth-glenfield" ], "talkSlugs": [ "how-to-hire-ai-engineers-when-everyone-is-cheating-with-ai" ] }, { "slug": "discord", "name": "Discord", "url": "https://ai.engineer/orgs/discord", "summary": "At AI Engineer World's Fair 2024, Ian Webster discussed how teams evaluated and iterated on Discord's Clyde, a conversational AI product operating at scale. His talk focused on reducing safety and reliability risks as developers tested and refined the application.", "speakerSlugs": [ "ian-webster" ], "talkSlugs": [ "iterating-on-llm-apps-at-scale-learnings-from-discord" ] }, { "slug": "docker", "name": "Docker", "url": "https://ai.engineer/orgs/docker", "summary": "At AI Engineer World's Fair 2026, Tushar Jain framed agent autonomy as a runtime safety problem. His talk examined how controls outside an agent's trust boundary can contain unexpected actions, with demonstrations spanning cloud execution, orchestrated agents, and Docker Sandboxes.", "speakerSlugs": [ "tushar-jain" ], "talkSlugs": [ "unlock-agent-autonomy-the-runtime-for-ai-native-systems" ] }, { "slug": "dspy", "name": "DSPy", "url": "https://ai.engineer/orgs/dspy", "summary": "At AI Engineer World's Fair 2026, DSPy's Maxime Rivest joined Isaac Miller to discuss separating task input-output signatures from models, prompts, and reasoning strategies. Their joint presentation demonstrated composable Python programs and evaluation-guided optimization while preserving stable task contracts.", "speakerSlugs": [ "maxime-rivest" ], "talkSlugs": [ "the-unreasonable-effectiveness-of-separating-the-task-from-the-model" ] }, { "slug": "duolingo", "name": "Duolingo", "url": "https://ai.engineer/orgs/duolingo", "summary": "At AI Engineer World's Fair 2026, Angel Ortmann Lee examined how automated cheating alerts can distort human judgment in remote proctoring. Her talk drew on a Duolingo English Test experiment involving fabricated alerts to question whether human oversight works when interfaces encourage approval.", "speakerSlugs": [ "angel-ortmann-lee" ], "talkSlugs": [ "build-ai-systems-for-discernment-not-approval-angel-ortmann-lee-duolingo" ] }, { "slug": "dx", "name": "DX", "url": "https://ai.engineer/orgs/dx", "summary": "At AI Engineer Code 2025, Justin Reock of DX examines how engineering leaders can evaluate AI-assisted work through software delivery, code quality, and business impact. His talk separates adoption mandates and perceived productivity from evidence of organizational improvement.", "speakerSlugs": [ "justin-reock" ], "talkSlugs": [ "leadership-in-ai-assisted-engineering" ] }, { "slug": "dylibso", "name": "Dylibso", "url": "https://ai.engineer/orgs/dylibso", "summary": "At AI Engineer World's Fair 2025, Dylibso-affiliated Benjamin Eckel joined Alex Volkov of Weights & Biases to discuss observability blind spots created by external Model Context Protocol tools. Their joint session demonstrated W&B Weave's Python-based MCP tracing and introduced Observable.tools.", "speakerSlugs": [ "benjamin-eckel" ], "talkSlugs": [ "the-state-of-mcp-observability-observable-tools-alex-volkov-and-benjamin-eckel-weights-biases-an" ] }, { "slug": "e2b", "name": "E2B", "url": "https://ai.engineer/orgs/e2b", "summary": "At AI Engineer World's Fair 2024, E2B’s Vasek Mlejnsky led a workshop building a Claude Artifacts-style application that securely executes model-generated code. The session combined Claude 3.5 Sonnet, the Vercel AI SDK, function calling, and E2B sandboxes, including Python data analysis and execution outputs.", "speakerSlugs": [ "vasek-mlejnsky" ], "talkSlugs": [ "how-to-add-secure-code-interpreting-in-your-ai-app" ] }, { "slug": "earendil", "name": "Earendil", "url": "https://ai.engineer/orgs/earendil", "summary": "At AI Engineer Europe 2026, Armin Ronacher and Cristina Poncela Cubeiro examined how frictionless AI-assisted coding can weaken human judgment. Their talk connects the momentum of generated code with skipped reviews and security or maintenance risks, distinguishing psychological dependence from architectural problems.", "speakerSlugs": [ "armin-ronacher", "cristina-poncela-cubeiro" ], "talkSlugs": [ "the-friction-is-your-judgment" ] }, { "slug": "echo-ai", "name": "Echo AI", "url": "https://ai.engineer/orgs/echo-ai", "summary": "Enterprise conversation intelligence connects Echo AI's Trey Doig and Log10's Arjun Bansal in a joint discussion at AI Engineer World's Fair 2024. The session covers ingesting customer interactions, configuring LLM pipelines, and extracting insights across conversation datasets.", "speakerSlugs": [ "trey-doig" ], "talkSlugs": [ "what-it-actually-takes-to-deploy-genai-applications-to-enterprises" ] }, { "slug": "effectful-technologies-inc", "name": "Effectful Technologies Inc", "url": "https://ai.engineer/orgs/effectful-technologies-inc", "summary": "At AI Engineer Europe 2026, Michael Arnaldi led a workshop on building reliable Effect applications with coding agents. Starting from an empty Bun/Vitest project, the session covered agent guidance, HTTP API patterns, edge validation, and Effect layer practices.", "speakerSlugs": [ "michael-arnaldi" ], "talkSlugs": [ "vibe-engineering-effect-apps" ] }, { "slug": "emergence", "name": "Emergence", "url": "https://ai.engineer/orgs/emergence", "summary": "At AI Engineer World's Fair 2024, Satya Nitta presented Emergence’s approach to automating complex business workflows. His talk introduced the Orchestrator, which coordinates LLMs and specialized agents to plan and verify work, alongside research into agent self-improvement and agent-oriented programming.", "speakerSlugs": [ "satya-nitta" ], "talkSlugs": [ "emergence-launch-ai-agents-and-the-future-enterprise" ] }, { "slug": "emulated", "name": "Emulated", "url": "https://ai.engineer/orgs/emulated", "summary": "At AI Engineer World's Fair 2026, Joseph Wang and Sid describe training autonomous software-engineering agents beyond repository-only coding tasks. Their talk examines simulated company workflows, production incidents, live traffic, and repeated deployments as environments for learning broader operational responsibilities.", "speakerSlugs": [ "joseph-wang", "sid" ], "talkSlugs": [ "emulated-the-data-for-fully-autonomous-software-engineers-and-companies" ] }, { "slug": "engram", "name": "Engram", "url": "https://ai.engineer/orgs/engram", "summary": "At AI Engineer World's Fair 2026, Engram's Jack Morris examines how pretrained language models can gain deep, personalized knowledge beyond public training data. His talk contrasts broad model knowledge with expertise in private information and rare skills, including AMD GPU kernel development.", "speakerSlugs": [ "jack-morris" ], "talkSlugs": [ "scaling-compute-on-context" ] }, { "slug": "ensemble-health-partners", "name": "Ensemble Health Partners", "url": "https://ai.engineer/orgs/ensemble-health-partners", "summary": "At AI Engineer World's Fair 2025, Ensemble Health Partners’ Nathan Wan presented healthcare revenue-cycle management as an applied AI opportunity. His talk examines how manual claims processes, payment delays, and denials strain hospitals, shifting attention from clinical applications to the operational work of securing payment.", "speakerSlugs": [ "nathan-wan" ], "talkSlugs": [ "ai-that-pays-lessons-from-revenue-cycle" ] }, { "slug": "entry-point-ai", "name": "Entry Point AI", "url": "https://ai.engineer/orgs/entry-point-ai", "summary": "At AI Engineer Summit 2023, Entry Point AI’s Mark Hennings presented no-code fine-tuning for practical business tasks. His talk compared fine-tuning with prompt engineering through considerations including prompt length, performance, cost, training datasets, and collaboration.", "speakerSlugs": [ "mark-hennings" ], "talkSlugs": [ "no-code-fine-tuning-mark-hennings" ] }, { "slug": "epicai-pro", "name": "EpicAI.pro", "url": "https://ai.engineer/orgs/epicai-pro", "summary": "At AI Engineer World's Fair 2025, EpicAI.pro's affiliated speaker Kent C. Dodds explains how Model Context Protocol connects AI assistants to applications, tools, and services. His talk pairs examples of multimodal interaction and dynamically generated interfaces with an authentication demonstration against an Epic Me server.", "speakerSlugs": [ "kent-c-dodds" ], "talkSlugs": [ "letting-ai-interface-with-your-app-with-mcp" ] }, { "slug": "erel-labs", "name": "Erel Labs", "url": "https://ai.engineer/orgs/erel-labs", "summary": "At AI Engineer Europe 2026, Erel Labs’ affiliated speaker Liad Yosef joined Ido Salomon to explain how MCP-UI evolved into MCP Apps. Their presentation examines branded, interactive application interfaces inside AI assistants and the potential for assistants to become application distribution channels.", "speakerSlugs": [ "liad-yosef" ], "talkSlugs": [ "mcp-ui-extending-the-frontier-liad-yosef-and-ido-salomon-mcp-apps" ] }, { "slug": "etsy", "name": "Etsy", "url": "https://ai.engineer/orgs/etsy", "summary": "At AI Engineer World's Fair 2026, Etsy-affiliated Aditya Bhargava examines how the surrounding harness shapes agent performance, especially with weaker models. His demonstration of Agency, a TypeScript-inspired agent programming language, develops that argument through concrete agent-building techniques.", "speakerSlugs": [ "aditya-bhargava" ], "talkSlugs": [ "what-if-the-harness-mattered-more-than-the-model-aditya-bhargava-etsy" ] }, { "slug": "exa", "name": "exa", "url": "https://ai.engineer/orgs/exa", "summary": "At AI Engineer World's Fair 2025, exa’s Will Bryk traced web search from PageRank and keyword matching to transformer-based retrieval for AI agents. His talk covered early model training, semantic search for complex research queries, and a coding demonstration of structured deep-research results.", "speakerSlugs": [ "will-bryk" ], "talkSlugs": [ "building-a-smarter-ai-agent-with-neural-rag" ] }, { "slug": "eyelevel-ai", "name": "EyeLevel.ai", "url": "https://ai.engineer/orgs/eyelevel-ai", "summary": "At AI Engineer World's Fair 2024, EyeLevel.ai's Benjamin Fletcher examined enterprise RAG failures caused by poor document ingestion, lost context, and irrelevant retrieval. His talk presented an alternative retrieval approach, illustrated through an Air France customer-support copilot.", "speakerSlugs": [ "benjamin-fletcher" ], "talkSlugs": [ "eyelevel-launch-your-rag-is-tripping-here-s-the-real-reason-why" ] }, { "slug": "fair-meta", "name": "FAIR, Meta", "url": "https://ai.engineer/orgs/fair-meta", "summary": "At AI Engineer Code 2025, Jacob Kahn presented Code World Model, a 32-billion-parameter dense research model that learns computational behavior from program execution traces and repository-level GitHub data. The talk connects world modeling to software-engineering agents that reason about code and modify repositories.", "speakerSlugs": [ "jacob-kahn" ], "talkSlugs": [ "code-world-model-building-world-models-for-computation" ] }, { "slug": "fal", "name": "Fal", "url": "https://ai.engineer/orgs/fal", "summary": "At AI Engineer World's Fair 2025, Gorkem Yurtseven surveyed generative media from computer-generated art and GANs to modern image and video models. His talk connected Fal's inference platform with applications in advertising, interactive campaigns, and e-commerce.", "speakerSlugs": [ "gorkem-yurtseven" ], "talkSlugs": [ "the-state-of-generative-media-today" ] }, { "slug": "fidelity-investments", "name": "Fidelity Investments", "url": "https://ai.engineer/orgs/fidelity-investments", "summary": "At AI Engineer World's Fair 2026, Fidelity Investments-affiliated speaker Sai Krishna Rallabandi presented Judith, a personal AI agent deployed in family-and-friends group chats. The talk examines privacy controls, role-aware routing, durable memory, and deciding when a shared agent should respond.", "speakerSlugs": [ "sai-krishna-rallabandi" ], "talkSlugs": [ "wearing-the-agent-engineering-a-family-and-friends-personal-agent-from-group-chats-to-glasses" ] }, { "slug": "filed", "name": "Filed", "url": "https://ai.engineer/orgs/filed", "summary": "At AI Engineer World's Fair 2026, Filed's Atul Ramachandran uses tax-preparation workflows to argue that chat and citations leave customers carrying verification and execution. He proposes agentic delegation, with users supervising AI workers and retaining control when intervention is needed.", "speakerSlugs": [ "atul-ramachandran" ], "talkSlugs": [ "chat-and-citations-won-t-save-your-vertical-ai" ] }, { "slug": "fixie-ai", "name": "Fixie.ai", "url": "https://ai.engineer/orgs/fixie-ai", "summary": "At AI Engineer Summit 2023, Matt Welsh introduces AI.JSX, an open-source TypeScript/JSX framework for reactive, composable LLM applications. His talk covers infrastructure concerns including vector databases, context windows, RAG, and tools, and closes with a low-latency conversational voice-agent demonstration.", "speakerSlugs": [ "matt-welsh" ], "talkSlugs": [ "building-reactive-ai-apps" ] }, { "slug": "flatfile", "name": "Flatfile", "url": "https://ai.engineer/orgs/flatfile", "summary": "At AI Engineer World's Fair 2025, Flatfile-affiliated Craig Wattrus presented AI coworkers as an interface design challenge. His talk organized interactions into invisible, ambient, inline, and conversational experiences, illustrated through personalized demos, agent-assisted data transformation, and collaborative agents.", "speakerSlugs": [ "craig-wattrus" ], "talkSlugs": [ "form-factors-for-your-new-ai-coworkers" ] }, { "slug": "flinn-ai", "name": "Flinn AI", "url": "https://ai.engineer/orgs/flinn-ai", "summary": "At AI Engineer Europe 2026, Flinn AI's Mardu Swanepoel examined design patterns for effective agents, including visible execution, personalization, and memory. The talk used examples from tools including Cursor, Claude Cowork, and Microsoft Word integration to make these patterns concrete.", "speakerSlugs": [ "mardu-swanepoel" ], "talkSlugs": [ "what-the-best-agents-share" ] }, { "slug": "flyerssoft", "name": "FlyersSoft", "url": "https://ai.engineer/orgs/flyerssoft", "summary": "At AI Engineer World's Fair 2026, Divakar Kumar of FlyersSoft presented an approach to adding AI agents to existing event-sourced financial systems for real-time fraud detection. The talk connects agent orchestration with the domain context needed to assess transactions.", "speakerSlugs": [ "divakar-kumar" ], "talkSlugs": [ "let-s-integrate-ai-agents-in-event-sourced-systems" ] }, { "slug": "forestwalk-labs", "name": "Forestwalk Labs", "url": "https://ai.engineer/orgs/forestwalk-labs", "summary": "At AI Engineer World's Fair 2026, Forestwalk Labs affiliated speaker Allen Pike presented agents that combine spoken input with rich visual output. His example was a meeting assistant that creates work items during conversation, connecting voice interaction to visible results.", "speakerSlugs": [ "allen-pike" ], "talkSlugs": [ "voice-in-visuals-out-the-agony-and-the-ecstasy" ] }, { "slug": "form3", "name": "Form3", "url": "https://ai.engineer/orgs/form3", "summary": "At AI Engineer World's Fair 2026, Form3's Moritz Johner presented PatchPilot, a production coding agent for CVE remediation across thousands of repositories. The talk addresses vulnerabilities that ordinary dependency updaters cannot resolve across base images, runtime versions, and CI failures.", "speakerSlugs": [ "moritz-johner" ], "talkSlugs": [ "we-gave-an-agent-production-code-access-and-then-tried-to-sleep-at-night" ] }, { "slug": "forward-future", "name": "Forward Future", "url": "https://ai.engineer/orgs/forward-future", "summary": "Local AI adoption is the focus of Forward Future's recorded presence at AI Engineer World's Fair 2026, where Matthew Berman joins a panel examining why adoption may be reaching an inflection point. The discussion spans distributed inference, computer vision, and simpler onboarding.", "speakerSlugs": [ "matthew-berman" ], "talkSlugs": [ "state-of-the-union-why-local-why-now" ] }, { "slug": "fractional-ai", "name": "Fractional AI", "url": "https://ai.engineer/orgs/fractional-ai", "summary": "At AI Engineer Summit 2025, Fractional AI's Eddie Siegel presented a production voice agent for qualitative research interviews. The talk describes how the OpenAI Realtime API, tool calls, and background text-based agents work together to monitor interview progress and select follow-up questions.", "speakerSlugs": [ "eddie-siegel" ], "talkSlugs": [ "voice-agents-the-good-the-bad-and-the-ugly" ] }, { "slug": "freeman-forrest", "name": "Freeman & Forrest", "url": "https://ai.engineer/orgs/freeman-forrest", "summary": "At AI Engineer World's Fair 2025, Freeman & Forrest's Forrest Brazeal joined Matt Ball of Augment Code to demonstrate investigating unfamiliar Apollo 11 Guidance Computer assembly code with Augment Agent. Their shared presentation moved from understanding legacy software to writing missing simulator code and completing a simulated lunar landing.", "speakerSlugs": [ "forrest-brazeal" ], "talkSlugs": [ "to-the-moon-navigating-deep-context-in-legacy-code-with-augment-agent" ] }, { "slug": "fujitsu-north-america", "name": "Fujitsu North America", "url": "https://ai.engineer/orgs/fujitsu-north-america", "summary": "At AI Engineer Europe 2026, Fujitsu North America's Dippu Kumar Singh presented a low-latency pipeline for contact-center voice intelligence. The talk connects multi-channel audio capture and speech-to-text with prompted LLM workflows that extract customer intent and operator actions, then synchronize verified results into CRM systems.", "speakerSlugs": [ "dippu-kumar-singh" ], "talkSlugs": [ "voiceops-fying-low-latency-intelligence-extraction-from-messy-audio-streams-dippu-kumar-singh" ] }, { "slug": "funstage-gmbh", "name": "Funstage GmbH", "url": "https://ai.engineer/orgs/funstage-gmbh", "summary": "At AI Engineer Code 2025, Alex Gavrilescu demonstrates Backlog.md, a terminal Kanban and Markdown-native task manager for humans and AI coding agents. His presentation shows how repository-backed descriptions, acceptance criteria, implementation plans, and human review checkpoints guide work with Claude Code.", "speakerSlugs": [ "alex-gavrilescu" ], "talkSlugs": [ "backlog-md-terminal-kanban-board-for-managing-tasks-with-ai-agents-alex-gavrilescu-funstage" ] }, { "slug": "g2i", "name": "G2i", "url": "https://ai.engineer/orgs/g2i", "summary": "At AI Engineer World's Fair 2026, G2i-affiliated speaker Ali Khial examined how software engineering benchmarks can misrepresent model capability. Drawing on SWE-Bench Pro and DeepSWE examples, his talk connected leaderboard distrust to benchmark quality and proposed human-authored, human-reviewed tasks grounded in real engineering value.", "speakerSlugs": [ "ali-khial" ], "talkSlugs": [ "benchmarks-the-good-the-bad-and-the-ugly" ] }, { "slug": "gabber", "name": "Gabber", "url": "https://ai.engineer/orgs/gabber", "summary": "Real-time voice AI deployment is the focus of Gabber’s AIE session, credited to Neil Dwyer and Jack Dwyer. Neil Dwyer explains deploying the open-source Orpheus text-to-speech model for affordable consumer applications, connecting voice demonstrations with the infrastructure needed to serve them.", "speakerSlugs": [ "neil-dwyer", "jack-dwyer" ], "talkSlugs": [ "serving-voice-ai-at-1-hr-open-source-loras-latency-load-balancing" ] }, { "slug": "galileo", "name": "Galileo", "url": "https://ai.engineer/orgs/galileo", "summary": "At AI Engineer World's Fair 2025, Galileo's affiliated speaker Jim Bennett examined why hallucinations, nondeterministic behavior, and multistep agent workflows call for observability-driven evaluation. A banking chatbot example showed how models can assess RAG outputs and diagnose missed tool calls.", "speakerSlugs": [ "jim-bennett" ], "talkSlugs": [ "taming-rogue-ai-agents-with-observability-driven-evaluation" ] }, { "slug": "gamma", "name": "Gamma", "url": "https://ai.engineer/orgs/gamma", "summary": "At AI Engineer World's Fair 2025, Gamma's Grant Lee discussed organizational design for AI startups, advocating adaptable multidisciplinary generalists and hands-on player-coach leaders. His talk questioned reflexive growth in management layers, while audience questions examined specialist hires, high agency, and work trials.", "speakerSlugs": [ "grant-lee" ], "talkSlugs": [ "rethinking-team-building-how-a-30-person-startup-serves-50-million-users-grant-lee-gamma" ] }, { "slug": "gas-town", "name": "Gas Town", "url": "https://ai.engineer/orgs/gas-town", "summary": "At AI Engineer World's Fair 2026, Gas Town's Steve Yegge argues that faster AI-assisted coding expands security exposure through familiar vulnerabilities and hallucinated package dependencies that attackers can exploit. Audience questions extend the discussion to agent credentials, autonomous actions, and prompt injection.", "speakerSlugs": [ "steve-yegge" ], "talkSlugs": [ "agentic-security-permissions-provenance-and-the-agent-supply-chain" ] }, { "slug": "gates-foundation", "name": "Gates Foundation", "url": "https://ai.engineer/orgs/gates-foundation", "summary": "At AI Engineer World's Fair 2026, Mike Phipps presented organizational processes and tacit knowledge as a durable advantage for enterprise AI. His talk described the Gates Foundation’s Strategic Intelligence Platform and its use of a semantic knowledge graph to support agentic workflows.", "speakerSlugs": [ "mike-phipps" ], "talkSlugs": [ "your-moat-is-your-data-model" ] }, { "slug": "genai-israel", "name": "GenAI Israel", "url": "https://ai.engineer/orgs/genai-israel", "summary": "At AI Engineer Summit 2025, GenAI Israel's Almog Baku presents the LLM Triangle in a prerecorded online talk. His framework connects foundation models, engineering techniques, and data through explicit standard operating procedures, addressing why proofs of concept fail with real users.", "speakerSlugs": [ "almog-baku" ], "talkSlugs": [ "the-llm-triangle-engineering-principles-for-robust-ai-applications" ] }, { "slug": "general-reasoning", "name": "General Reasoning", "url": "https://ai.engineer/orgs/general-reasoning", "summary": "At AI Engineer World's Fair 2026, Ross Taylor and Chengxi Taylor discuss reinforcement learning, training environments, compute, and objectives for agents tackling long-horizon tasks. Ross connects lessons from model post-training to RLHF's role in making capable base models useful products.", "speakerSlugs": [ "ross-taylor", "chengxi-taylor" ], "talkSlugs": [ "scaling-to-long-horizons" ] }, { "slug": "gensx", "name": "GenSX", "url": "https://ai.engineer/orgs/gensx", "summary": "At AI Engineer World's Fair 2025, GenSX's Evan Boyle examined how slow, unreliable LLM calls and agent workflows lasting minutes or hours break conventional web application assumptions. His talk compared queues, Airflow, and Temporal and demonstrated background scraping and content generation.", "speakerSlugs": [ "evan-boyle" ], "talkSlugs": [ "how-agents-broke-app-level-infrastructure" ] }, { "slug": "gimlet-labs", "name": "Gimlet Labs", "url": "https://ai.engineer/orgs/gimlet-labs", "summary": "At AI Engineer Code 2025, Gimlet Labs’ Natalie Serrino presented AI agents that generate and optimize GPU kernels for PyTorch workloads across heterogeneous hardware. Her talk paired an H100-targeted workflow with Apple M4/Metal results on KernelBench v0.1.", "speakerSlugs": [ "natalie-serrino" ], "talkSlugs": [ "ai-kernel-generation-what-s-working-what-s-not-what-s-next" ] }, { "slug": "glean", "name": "Glean", "url": "https://ai.engineer/orgs/glean", "summary": "At AI Engineer World's Fair 2025, Glean's Chau Tran examined how enterprise agents can follow organization-specific procedures. The talk contrasts explicitly orchestrated workflows with LLM-directed agents, weighing control, debugging, latency and cost, and argues for evaluating execution steps alongside final outputs.", "speakerSlugs": [ "chau-tran" ], "talkSlugs": [ "how-to-build-enterprise-aware-agents" ] }, { "slug": "glow", "name": "Glow", "url": "https://ai.engineer/orgs/glow", "summary": "David Vorick's Glow presentation addressed solar deployment within the HF0 Residency startup showcase at AI Engineer World's Fair 2025. The shared session placed that subject alongside pitches on creative tools, enterprise data, voice models, and model routing.", "speakerSlugs": [ "david-vorick" ], "talkSlugs": [ "the-next-unicorns-7-top-ai-startups-from-the-hf0-residency" ] }, { "slug": "good-collective", "name": "Good Collective", "url": "https://ai.engineer/orgs/good-collective", "summary": "Graph construction, retrieval and code analysis anchor Good Collective's recorded AIE appearance with Tim Ainge. At AI Engineer World's Fair 2026, he examines where graph-native techniques help AI applications, using legal citations and code-context retrieval without presenting GraphRAG as a universal solution.", "speakerSlugs": [ "tim-ainge" ], "talkSlugs": [ "a-practitioner-s-guide-to-graphs-tim-ainge-good-collective" ] }, { "slug": "goodfire", "name": "Goodfire", "url": "https://ai.engineer/orgs/goodfire", "summary": "At AI Engineer World's Fair 2025, Goodfire’s Mark Bissell presented mechanistic interpretability as a way to inspect and steer neural networks through meaningful internal features. His examples connected model internals to privacy controls, multilingual PII detection, and image interfaces.", "speakerSlugs": [ "mark-bissell" ], "talkSlugs": [ "why-you-should-care-about-ai-interpretability" ] }, { "slug": "google-youtube", "name": "Google / YouTube", "url": "https://ai.engineer/orgs/google-youtube", "summary": "At AI Engineer World's Fair 2025, Devansh Tandon describes adapting Gemini checkpoints into a Large Recommender Model for YouTube video retrieval and ranking. The talk covers personalized generative retrieval and production deployment, with audience questions on cold starts, visual encoders, and pre-training data.", "speakerSlugs": [ "devansh-tandon" ], "talkSlugs": [ "teaching-gemini-to-speak-youtube-adapting-llms-for-video-recommendations-to-2b-dau" ] }, { "slug": "google-youtube-ads", "name": "Google / YouTube Ads", "url": "https://ai.engineer/orgs/google-youtube-ads", "summary": "At AI Engineer World's Fair 2026, Preetika Bhateja and Daniel Bump discussed improving agent behavior through LLM-friendly tools, iterative evaluations, and feedback. Their presentation covered inspecting agent reasoning, tracking recurring failures, and selecting metrics for launch readiness.", "speakerSlugs": [ "preetika-bhateja", "daniel-bump" ], "talkSlugs": [ "how-evals-and-prompts-shape-agent-behavior-preetika-bhateja-daniel-bump-youtube-ads" ] }, { "slug": "google-photos", "name": "Google Photos", "url": "https://ai.engineer/orgs/google-photos", "summary": "At AI Engineer World's Fair 2025, Google Photos speaker Kelvin Ma explained how computational photography and generative AI become production-ready mobile editing features. His talk connects a shared cross-platform C++ stack and on-device inference with model-size constraints, evaluation, and unpredictable edge cases.", "speakerSlugs": [ "kelvin-ma" ], "talkSlugs": [ "magic-editor-under-the-hood-weaving-generative-ai-into-a-billion-user-app" ] }, { "slug": "gradient", "name": "Gradient", "url": "https://ai.engineer/orgs/gradient", "summary": "At AI Engineer World's Fair 2024, Gradient’s Leo Pekelis presented Albatross, a finance-domain language model adapted from Llama 2. His talk covers domain-specific training data, human review, and synthetic augmentation, alongside reported improvements on finance evaluations and NVIDIA's RULER long-context benchmark.", "speakerSlugs": [ "leo-pekelis" ], "talkSlugs": [ "training-albatross-an-expert-finance-llm" ] }, { "slug": "gradium-ai", "name": "Gradium AI", "url": "https://ai.engineer/orgs/gradium-ai", "summary": "At AI Engineer Europe 2026, Neil Zeghidour examined why voice agents still fall short of natural conversation. His talk contrasted cascaded speech and language systems with speech-to-speech dialogue, using demonstrations to explain simultaneous listening and speaking, Moshi’s full-duplex architecture, and paralinguistic understanding.", "speakerSlugs": [ "neil-zeghidour" ], "talkSlugs": [ "neil-zeghidour-voice-ai-when-is-the-her-moment" ] }, { "slug": "granola", "name": "Granola", "url": "https://ai.engineer/orgs/granola", "summary": "At AI Engineer Europe 2026, Mehedi Hassan described building AI meeting-chat features at Granola through continuous feedback loops. His talk connects challenges in web search and role-specific outputs with practical ways to accelerate experimentation, including pull-request previews and Cursor-assisted testing.", "speakerSlugs": [ "mehedi-hassan" ], "talkSlugs": [ "feedback-loops-are-all-you-need" ] }, { "slug": "grit", "name": "Grit", "url": "https://ai.engineer/orgs/grit", "summary": "At AI Engineer World's Fair 2024, Grit's Morgante Pell discusses AI coding agents for maintaining and migrating existing enterprise codebases. He uses OpenTelemetry to illustrate organizational migration challenges and describes coordinating large-scale pull-request workflows for expert engineers.", "speakerSlugs": [ "morgante-pell" ], "talkSlugs": [ "code-generation-and-maintenance-at-scale" ] }, { "slug": "groq", "name": "Groq", "url": "https://ai.engineer/orgs/groq", "summary": "At AI Engineer World's Fair 2024, Groq’s Sunny Madra argued that faster, cheaper token processing could reshape software and creative work. His talk connected inference performance with potential research, multimodal, enterprise, and security applications.", "speakerSlugs": [ "sunny-madra" ], "talkSlugs": [ "breaking-ai-s-1-gigahertz-barrier" ] }, { "slug": "growth-cyber", "name": "Growth Cyber", "url": "https://ai.engineer/orgs/growth-cyber", "summary": "At AI Engineer World's Fair 2025, Growth Cyber's Allie Howe presented trustworthy AI as a combination of security and safety. Her talk connects shared responsibility across product, engineering, and security teams with practical defenses against misuse, data leakage, prompt injection, and harmful outputs.", "speakerSlugs": [ "allie-howe" ], "talkSlugs": [ "how-to-build-trustworthy-ai" ] }, { "slug": "gru-ai", "name": "Gru.ai", "url": "https://ai.engineer/orgs/gru-ai", "summary": "At AI Engineer Summit 2025, Hailong Zhang contrasts interactive coding assistants with asynchronous agents that complete repository tasks. His Test Gru demonstration follows a workflow from detecting pull-request changes to generating and running unit tests and submitting reviewed changes.", "speakerSlugs": [ "hailong-zhang" ], "talkSlugs": [ "how-coding-agents-change-software-development-forever-hailong-zhang" ] }, { "slug": "guardrails-ai", "name": "Guardrails AI", "url": "https://ai.engineer/orgs/guardrails-ai", "summary": "At AI Engineer Summit 2023, Shreya Rajpal examined why generative AI prototypes can fail in production. Her talk connects stochastic LLM outputs to structural, correctness, and security requirements, presenting a trust-but-verify programming model for checking results before accepting them.", "speakerSlugs": [ "shreya-rajpal" ], "talkSlugs": [ "trust-but-verify" ] }, { "slug": "gumloop", "name": "Gumloop", "url": "https://ai.engineer/orgs/gumloop", "summary": "At AI Engineer World's Fair 2025, Max Brodeur-Urbas described Gumloop’s efforts to scale an AI workflow-automation company with a small team. His talk covered selective hiring, recruiting from the customer community, and using Gumloop to analyze chatbot conversations and automate internal work.", "speakerSlugs": [ "max-brodeur-urbas" ], "talkSlugs": [ "building-a-10-person-unicorn" ] }, { "slug": "haize-labs", "name": "Haize Labs", "url": "https://ai.engineer/orgs/haize-labs", "summary": "At AI Engineer World's Fair 2025, Leonard Tang of Haize Labs presented Haizing, an approach to fuzzing and property-based testing for generative AI. His talk argues that static golden-dataset evaluations miss brittleness that can prevent applications from reaching production.", "speakerSlugs": [ "leonard-tang" ], "talkSlugs": [ "fuzzing-in-the-genai-era" ] }, { "slug": "halluminate", "name": "Halluminate", "url": "https://ai.engineer/orgs/halluminate", "summary": "At AI Engineer World's Fair 2025, Halluminate’s Jerry Wu and Wyatt Marshall examined how browser agents operate and used WebBench to assess information retrieval and action-taking workflows. Their talk connected deployment choices to practical constraints, including authentication, CAPTCHAs, latency, and browser infrastructure.", "speakerSlugs": [ "jerry-wu", "wyatt-marshall" ], "talkSlugs": [ "the-current-state-of-browser-agents" ] }, { "slug": "harvey", "name": "Harvey", "url": "https://ai.engineer/orgs/harvey", "summary": "Enterprise legal RAG is the focus of Harvey's recorded appearance with Calvin Qi at AI Engineer World's Fair 2025. In a joint session with Chang She of LanceDB, the discussion covers complex regulatory questions, domain-specific evaluation, confidentiality, customer isolation, and retention requirements.", "speakerSlugs": [ "calvin-qi" ], "talkSlugs": [ "scaling-enterprise-grade-rag-systems-lessons-from-the-legal-frontier" ] }, { "slug": "heroku", "name": "Heroku", "url": "https://ai.engineer/orgs/heroku", "summary": "Julián Duque and Anush DSouza’s workshop at AI Engineer World's Fair 2025 covers building agentic applications with Heroku Managed Inference and Agents. Participants deploy a Jupyter Notebook, inspect database schemas, generate SQL, and integrate custom MCP tools including Brave Search.", "speakerSlugs": [ "julian-duque", "anush-dsouza" ], "talkSlugs": [ "building-agentic-applications-with-heroku-managed-inference-and-agents-julian-duque-and-anush-ds" ] }, { "slug": "hey-ai", "name": "Hey AI", "url": "https://ai.engineer/orgs/hey-ai", "summary": "At AI Engineer World's Fair 2026, Veronica Hylak of Hey AI addressed how founders can make agentic-AI products understandable to customers. Her talk framed product storytelling around recognizable customer problems and specific workflow changes, with examples spanning security operations and AI-agent observability.", "speakerSlugs": [ "veronica-hylak" ], "talkSlugs": [ "your-ai-product-will-fail-unless-you-can-explain-it" ] }, { "slug": "heygen", "name": "HeyGen", "url": "https://ai.engineer/orgs/heygen", "summary": "At AI Engineer World's Fair 2026, HeyGen-affiliated James Russo presented HyperFrames and the case for agents creating videos with familiar HTML, CSS, and JavaScript. His talk contrasted this approach with traditional creative tools and Remotion, while emphasizing creative taste and iterative human editing.", "speakerSlugs": [ "james-russo" ], "talkSlugs": [ "html-is-all-agents-need" ] }, { "slug": "hiddenlayer", "name": "HiddenLayer", "url": "https://ai.engineer/orgs/hiddenlayer", "summary": "At AI Engineer World's Fair 2024, HiddenLayer's affiliated speaker Dr. Andrew Davis addressed practical security for machine-learning and multimodal AI systems. His talk covers adversarial examples, production API monitoring, software vulnerabilities, data poisoning, and prompt injection against email-connected agents.", "speakerSlugs": [ "dr-andrew-davis" ], "talkSlugs": [ "building-security-around-ml" ] }, { "slug": "higharc", "name": "Higharc", "url": "https://ai.engineer/orgs/higharc", "summary": "At AI Engineer World's Fair 2026, Higharc's Vaidas Razgaitis described turning frontier machine-learning research into production homebuilding features. His talk connects technical design documents covering domain context and business goals with the architecture and delivery practices needed to ship those features.", "speakerSlugs": [ "vaidas-razgaitis" ], "talkSlugs": [ "research-to-reality-bringing-frontier-ml-research-to-production" ] }, { "slug": "hinge-health", "name": "Hinge Health", "url": "https://ai.engineer/orgs/hinge-health", "summary": "At AI Engineer World's Fair 2026, Rashi Agrawal presented architectural safeguards for member-facing healthcare AI. Her talk covers removing protected health information at ingestion, isolating production systems, and enforcing emergency escalation and clinical routing in deterministic code before model execution.", "speakerSlugs": [ "rashi-agrawal" ], "talkSlugs": [ "guardrails-first-engineering-member-facing-health-ai" ] }, { "slug": "hippocratic-ai", "name": "Hippocratic AI", "url": "https://ai.engineer/orgs/hippocratic-ai", "summary": "At AI Engineer World's Fair 2026, Vivek Muppalla presented healthcare voice agents for proactive patient outreach without replacing clinicians. His talk connects medication-sensitive conversations and nurse escalation with Polaris, a conversational architecture supported by specialist models and tool-call verifiers.", "speakerSlugs": [ "vivek-muppalla" ], "talkSlugs": [ "200-million-patient-interactions-later-what-the-generic-voice-stack-misses" ] }, { "slug": "honeyhive", "name": "HoneyHive", "url": "https://ai.engineer/orgs/honeyhive", "summary": "At AI Engineer Summit 2025, HoneyHive's Mohak Sharma examined why developer-written test sets and generic metrics can misrepresent production performance. His talk recommends domain-expert-defined criteria and adding problematic production queries to evaluation datasets so tests reflect actual failures.", "speakerSlugs": [ "mohak-sharma" ], "talkSlugs": [ "your-evals-are-meaningless-and-here-s-how-to-fix-them" ] }, { "slug": "hud", "name": "Hud", "url": "https://ai.engineer/orgs/hud", "summary": "At AI Engineer World's Fair 2026, May Walter presented a workflow for finding and prioritizing performance improvements using function-level runtime intelligence. The talk connects GitHub Actions, Claude Code, and production context accessible through MCP to repository analysis and actionable Slack reports.", "speakerSlugs": [ "may-walter" ], "talkSlugs": [ "from-blind-spots-to-merged-prs-continuous-agentic-performance-optimization" ] }, { "slug": "huge", "name": "Huge", "url": "https://ai.engineer/orgs/huge", "summary": "At AI Engineer World's Fair 2025, Huge's Alex Liss presented AI simulation as a way to improve interface design and usability evaluation. His talk connects audience data, intent mapping, and simulated Intelligent Twins, with a sports-website audit illustrating task completion and friction discovery.", "speakerSlugs": [ "alex-liss" ], "talkSlugs": [ "invisible-users-invisible-interfaces-accelerating-design-iteration-with-ai-simulation" ] }, { "slug": "humanloop", "name": "Humanloop", "url": "https://ai.engineer/orgs/humanloop", "summary": "Raza Habib's appearance at AI Engineer World's Fair 2024 examines how enterprises generate revenue and reduce costs with production LLM applications. His talk connects examples including GitHub Copilot, Duolingo, Filevine, and Ironclad with evaluation-first development and collaboration between product engineers and domain experts.", "speakerSlugs": [ "raza-habib" ], "talkSlugs": [ "real-roi-lessons-from-enterprises-that-have-already-succeeded-with-llms-redacted-username-scale" ] }, { "slug": "huxe", "name": "Huxe", "url": "https://ai.engineer/orgs/huxe", "summary": "At AI Engineer World's Fair 2025, Huxe-affiliated speaker Raiza Martin discussed how AI blurs the boundaries between product, engineering, and design. Her talk frames that shift as an opportunity to rebuild interfaces around concrete user needs.", "speakerSlugs": [ "raiza-martin" ], "talkSlugs": [ "everything-is-ugly-so-go-build-something-that-isn-t" ] }, { "slug": "hyperbolic", "name": "Hyperbolic", "url": "https://ai.engineer/orgs/hyperbolic", "summary": "At AI Engineer World's Fair 2025, Dr. Jasper Zhang argued that additional data centers alone cannot meet AI compute demand while existing GPUs remain underused. His talk presented Hyperbolic’s approach to pooling fragmented capacity through a GPU marketplace with flexible rentals.", "speakerSlugs": [ "dr-jasper-zhang" ], "talkSlugs": [ "why-we-don-t-need-more-data-centers" ] }, { "slug": "hyperspace", "name": "Hyperspace", "url": "https://ai.engineer/orgs/hyperspace", "summary": "At AI Engineer World's Fair 2024, Nicolas Schlaepfer presented Hyperspace's decentralized AI inference network and AI OS desktop product. His demonstration turned user requests into editable directed acyclic graphs, with HyperEngine V3 streaming workflow nodes into a React Flow editor.", "speakerSlugs": [ "nicolas-schlaepfer" ], "talkSlugs": [ "hyperspace-more-nodes-is-all-you-need" ] }, { "slug": "ikea", "name": "IKEA", "url": "https://ai.engineer/orgs/ikea", "summary": "At AI Engineer Europe 2026, IKEA-affiliated speaker Raj Navakoti presented Demand-Driven Context: building enterprise agent knowledge bases from real tasks, observed failures, and missing domain knowledge. The workshop examined how fragmented information and stale documentation complicate agents’ work.", "speakerSlugs": [ "raj-navakoti" ], "talkSlugs": [ "build-your-first-demand-driven-context-base-let-ai-agents-tell-you-what-they-need" ] }, { "slug": "imbue", "name": "Imbue", "url": "https://ai.engineer/orgs/imbue", "summary": "At AI Engineer World's Fair 2025, Josh Albrecht introduced Sculptor, an experimental coding-agent environment for making AI-generated software trustworthy in established production codebases. His talk connects research, planning, specifications, and strict style guides with immediate AI-assisted review.", "speakerSlugs": [ "josh-albrecht" ], "talkSlugs": [ "beyond-the-prototype-using-ai-to-write-high-quality-code" ] }, { "slug": "incident-io", "name": "incident.io", "url": "https://ai.engineer/orgs/incident-io", "summary": "At AI Engineer Europe 2026, Lawrence Jones described how incident.io uses coding agents to debug complex AI SRE systems. His talk covers an agent-friendly CLI for oversized evaluation suites and filesystem interfaces that expose debugging tools and prompt hierarchies to Claude Code.", "speakerSlugs": [ "lawrence-jones" ], "talkSlugs": [ "lawrence-jones-fighting-ai-with-ai" ] }, { "slug": "incubator-for-artificial-intelligence-i-ai", "name": "Incubator for Artificial Intelligence (i.AI)", "url": "https://ai.engineer/orgs/incubator-for-artificial-intelligence-i-ai", "summary": "At AI Engineer World's Fair 2025, James Lowe presented AI product management through government consultation analysis and ministerial information workflows. His talk encouraged engineers to take on product leadership, combining technical expertise with experimentation and validation with real users.", "speakerSlugs": [ "james-lowe" ], "talkSlugs": [ "why-your-product-needs-an-ai-product-manager-and-why-it-should-be-you" ] }, { "slug": "independent-state-of-data", "name": "Independent / State of Data", "url": "https://ai.engineer/orgs/independent-state-of-data", "summary": "At AI Engineer World's Fair 2026, Sean Cai examined the shift in AI training-data markets from static annotations and final-state records toward authentic enterprise workflows. His State of Data talk connected application maturity to whether tasks can be reliably verified.", "speakerSlugs": [ "sean-cai" ], "talkSlugs": [ "state-of-data" ] }, { "slug": "inngest", "name": "Inngest", "url": "https://ai.engineer/orgs/inngest", "summary": "At AI Engineer World's Fair 2026, Dan Farrelly of Inngest discusses building agent architectures around stable execution primitives as models and frameworks change. His talk covers durable, resumable execution, human-in-the-loop and sub-agent orchestration, sandboxing, and end-to-end observability.", "speakerSlugs": [ "dan-farrelly" ], "talkSlugs": [ "your-agent-architecture-has-a-half-life-of-6-months" ] }, { "slug": "insight-sciences", "name": "Insight Sciences", "url": "https://ai.engineer/orgs/insight-sciences", "summary": "At AI Engineer World's Fair 2026, Insight Sciences' Ishan Anand examines language models as simulated research respondents. He frames synthetic personas as forecasts rather than independent human observations, examining how missing context, prompt sensitivity, and persona construction can distort predictions of human preferences.", "speakerSlugs": [ "ishan-anand" ], "talkSlugs": [ "persona-engineering-a-field-guide-to-ai-synthetic-personas-ishan-anand-insightsciences-ai" ] }, { "slug": "instacart", "name": "Instacart", "url": "https://ai.engineer/orgs/instacart", "summary": "Grocery search and product discovery anchor Vinesh Gudla's recorded Instacart appearance at AI Engineer World's Fair 2025. The talk describes using LLMs to ground category predictions in taxonomy and conversion data and generate substitute and synonymous query rewrites.", "speakerSlugs": [ "vinesh-gudla" ], "talkSlugs": [ "how-instacart-transformed-its-search-and-discovery-using-an-llm-driven-approach" ] }, { "slug": "ionic-commerce", "name": "Ionic Commerce", "url": "https://ai.engineer/orgs/ionic-commerce", "summary": "At AI Engineer World's Fair 2024, Justin McCarty presented Ionic Commerce infrastructure for agents to discover products, make personalized purchasing decisions, and complete transactions. The talk connects these steps through a unified commerce API and payment tokens.", "speakerSlugs": [ "justin-mccarty" ], "talkSlugs": [ "ionic-launch-opening-the-economy-to-ai-agents" ] }, { "slug": "isadora-co-the-bloom-house-ai", "name": "Isadora & Co | The Bloom House AI", "url": "https://ai.engineer/orgs/isadora-co-the-bloom-house-ai", "summary": "At AI Engineer World's Fair 2026, Isadora Martin-Dye presented an architecture for dependable AI brand voice, separating identity rules, situational context, voice examples, and output checks. Wedding-venue assistants and a missing-person tool illustrate failures involving invented facts, insensitive responses, and misleading identity claims.", "speakerSlugs": [ "isadora-martin-dye" ], "talkSlugs": [ "stop-writing-tone-instructions-layer-them" ] }, { "slug": "it-revolution", "name": "IT Revolution", "url": "https://ai.engineer/orgs/it-revolution", "summary": "At AI Engineer Code 2025, Gene Kim joined Steve Yegge to discuss a possible shift from traditional IDEs to AI-native development environments and orchestrated coding agents. Kim connected vibe coding to DevOps and the organizational changes surrounding AI-assisted software development.", "speakerSlugs": [ "gene-kim" ], "talkSlugs": [ "2026-the-year-the-ide-died" ] }, { "slug": "iterate", "name": "Iterate", "url": "https://ai.engineer/orgs/iterate", "summary": "At AI Engineer Europe 2026, Misha Kaletsky and Jonas Templestein demonstrate an extensible agent harness built around programmable event streams and stream processors. Their workshop connects the design to runnable cURL examples and a TypeScript SDK installed with pnpm.", "speakerSlugs": [ "misha-kaletsky", "jonas-templestein" ], "talkSlugs": [ "make-your-own-event-sourced-agent-harness-using-stream-processors" ] }, { "slug": "jam", "name": "Jam", "url": "https://ai.engineer/orgs/jam", "summary": "At AI Engineer World's Fair 2025, Jam's Dani Grant joined Chelcie Taylor of Notable Capital for an interactive fundraising workshop for AI startup founders. Their discussion covered investor psychology, warm introductions, cold outreach, and differentiated AI pitches.", "speakerSlugs": [ "dani-grant" ], "talkSlugs": [ "the-ai-engineer-s-guide-to-raising-vc-dani-grant-jam-chelcie-taylor-notable-capital" ] }, { "slug": "jane-street", "name": "Jane Street", "url": "https://ai.engineer/orgs/jane-street", "summary": "At AI Engineer Summit 2025, John Crepezzi described building LLM-powered developer tools for Jane Street's OCaml-centric environment. Scarce public training data, proprietary infrastructure, a Mercurial monorepo, and widespread Emacs usage frame the talk's explanation of why off-the-shelf assistants face limits.", "speakerSlugs": [ "john-crepezzi" ], "talkSlugs": [ "building-ai-powered-developer-tools-at-jane-street" ] }, { "slug": "jedi", "name": "Jedi", "url": "https://ai.engineer/orgs/jedi", "summary": "At AI Engineer Summit 2025, Jedi-affiliated Patrick Debois applied Reverse Conway's Law to how generative AI agents could reshape organizational structures. His talk traced a possible progression from coding copilots to domain-specific teammates and potential managers, with implications for engineering roles and team composition.", "speakerSlugs": [ "patrick-debois" ], "talkSlugs": [ "reverse-conway-s-law-and-genai-how-agents-will-take-over-the-organisation" ] }, { "slug": "jellyfish", "name": "Jellyfish", "url": "https://ai.engineer/orgs/jellyfish", "summary": "At AI Engineer Code 2025, Jellyfish’s Nicholas Arcolano examined how engineering teams adopt AI coding tools using observed developer behavior. His talk connects adoption with pull-request throughput, cycle time, and larger changes, while distinguishing coding-tool usage from fully autonomous contributions.", "speakerSlugs": [ "nicholas-arcolano" ], "talkSlugs": [ "what-data-from-20-million-pull-requests-reveal-about-ai-transformation" ] }, { "slug": "joinin-ai", "name": "JoinIn AI", "url": "https://ai.engineer/orgs/joinin-ai", "summary": "At AI Engineer World's Fair 2026, JoinIn AI's Ted Johnson argues that prompting preserves the punch-card pattern: assemble a complete request, submit it, wait, and retry. His talk contrasts advancing language-model capabilities with interaction designs that still require people to accommodate machines.", "speakerSlugs": [ "ted-johnson" ], "talkSlugs": [ "the-prompt-is-still-a-punch-card" ] }, { "slug": "jointly", "name": "Jointly", "url": "https://ai.engineer/orgs/jointly", "summary": "At AI Engineer Code 2025, Alberto Romero of Jointly presented Meta-ACE, a framework for optimizing agents across context, compute, verification, memory, and parameter adaptation. The talk examined task profiling and a learned meta-controller for allocating resources to different optimization strategies.", "speakerSlugs": [ "alberto-romero" ], "talkSlugs": [ "the-unbearable-lightness-of-agent-optimization" ] }, { "slug": "jp-morgan-chase", "name": "JP Morgan Chase", "url": "https://ai.engineer/orgs/jp-morgan-chase", "summary": "At AI Engineer World's Fair 2026, Ritvik Pandya described representing payment API requests as short-lived, learned execution DAGs. The talk connects anomalous processing paths with node-level failure localization and distinguishes isolated anomalies from longer-term drift.", "speakerSlugs": [ "ritvik-pandya" ], "talkSlugs": [ "learned-execution-graphs-for-anomaly-detection-drift-in-apis-ritvik-pandya-jp-morgan-chase" ] }, { "slug": "k-scale-labs", "name": "K-Scale Labs", "url": "https://ai.engineer/orgs/k-scale-labs", "summary": "At AI Engineer World's Fair 2025, Jingxiang \"JX\" Mo presented K-Scale Labs’ K-Bot, an approximately $9,000 open-source humanoid, alongside the smaller Z-Bot. The talk covered openly available hardware and software, modular upgrades, and Python/Rust tools for developers.", "speakerSlugs": [ "jingxiang-jx-mo" ], "talkSlugs": [ "your-personal-open-source-humanoid-robot-for-8-999-jingxiang-jx-mo-k-scale-labs" ] }, { "slug": "kalmantic-labs", "name": "Kalmantic Labs", "url": "https://ai.engineer/orgs/kalmantic-labs", "summary": "At AI Engineer World's Fair 2026, Thiyagarajan Maruthavanan examined unpredictable inference costs and token waste in agent loops. Drawing on his Ultasono application and a migration from Anthropic to NVIDIA DGX Spark hardware, he compared rented inference, cloud token factories, and owned local infrastructure.", "speakerSlugs": [ "thiyagarajan-maruthavanan" ], "talkSlugs": [ "stop-renting-your-cognitive-infrastructure" ] }, { "slug": "khan-academy", "name": "Khan Academy", "url": "https://ai.engineer/orgs/khan-academy", "summary": "At AI Engineer World's Fair 2024, Khan Academy's Shawn Jansepar presented Khanmigo's approach to personalized tutoring and teacher assistance. His case study covers Socratic questioning, mathematical accuracy, context-aware learning, and classroom interfaces, including iterative essay coaching and teacher visibility into student work.", "speakerSlugs": [ "shawn-jansepar" ], "talkSlugs": [ "scaling-ai-in-education-a-khanmigo-case-study" ] }, { "slug": "kilo-code", "name": "Kilo Code", "url": "https://ai.engineer/orgs/kilo-code", "summary": "At AI Engineer Europe 2026, Brendan O'Leary presented agentic engineering as actively directing AI coding agents. His talk emphasized treating agents as capable but fallible junior developers, with people retaining control through frequent Git commits, code review, build checks, and tests.", "speakerSlugs": [ "brendan-o-leary" ], "talkSlugs": [ "agentic-engineering-working-with-ai-not-just-using-it-brendan-o-leary" ] }, { "slug": "klarity", "name": "Klarity", "url": "https://ai.engineer/orgs/klarity", "summary": "At AI Engineer World's Fair 2024, Klarity's Nischal Nadhamuni joined Sheila Gulati to discuss evaluating agentic AI against operational objectives, human values, and real-world performance. Nadhamuni connected enterprise evaluation to document-heavy finance and accounting automation.", "speakerSlugs": [ "nischal-nadhamuni" ], "talkSlugs": [ "e-values-evaluating-the-values-of-ai" ] }, { "slug": "krafton", "name": "KRAFTON", "url": "https://ai.engineer/orgs/krafton", "summary": "At AI Engineer World's Fair 2026, KRAFTON-affiliated speaker Kyle Jaejun Lee described running a personal fleet of coding agents across a MacBook and Linux machines. His talk examines delegated agent roles, disk-based handoffs, and failures involving memory exhaustion, laptop interruptions, and divergent state.", "speakerSlugs": [ "kyle-jaejun-lee" ], "talkSlugs": [ "i-run-a-fleet-of-ai-agents-across-three-machines-here-s-what-broke" ] }, { "slug": "lastmile-ai", "name": "LastMile AI", "url": "https://ai.engineer/orgs/lastmile-ai", "summary": "At AI Engineer World's Fair 2025, LastMile AI's recorded speaker Sarmad Qadri explains how mcp-agent exposes AI agents as Model Context Protocol servers. His talk describes how compatible clients can invoke and compose those agents through a common interface.", "speakerSlugs": [ "sarmad-qadri" ], "talkSlugs": [ "exposing-agents-as-mcp-servers-with-mcp-agent-sarmad-qadri" ] }, { "slug": "latchbio", "name": "LatchBio", "url": "https://ai.engineer/orgs/latchbio", "summary": "At AI Engineer World's Fair 2026, LatchBio's Kenny Workman presented SpatialBench for evaluating scientific AI agents on spatial-analysis tasks. He also introduced SpatialBench-Long, extending evaluation to realistic, multistep biological investigations as single-cell and spatial-biology datasets grow.", "speakerSlugs": [ "kenny-workman" ], "talkSlugs": [ "verifiable-environments-for-ai-in-biology-kenny-workman-latchbio" ] }, { "slug": "latent-space-university", "name": "Latent Space University", "url": "https://ai.engineer/orgs/latent-space-university", "summary": "At AI Engineer Summit 2023, Noah Hein led a hands-on introduction to AI engineering using Latent Space University material. Participants configured a Telegram bot and development environment, connected to the OpenAI API, and discussed AI-assisted code generation and image-generation applications including RoomGPT.", "speakerSlugs": [ "noah-hein" ], "talkSlugs": [ "ai-engineering-101" ] }, { "slug": "laude-institute", "name": "Laude Institute", "url": "https://ai.engineer/orgs/laude-institute", "summary": "At AI Engineer World's Fair 2026, Alex Shaw presents agent improvement as an empirical process built around evaluations and rollouts. The talk introduces Harbor for sandboxed evaluations and reinforcement-learning environments, with a demonstration of parallel Terminal-Bench execution using Codex and GPT-5.5.", "speakerSlugs": [ "alex-shaw", "ryan-marten" ], "talkSlugs": [ "everything-is-a-rollout-alex-shaw-ryan-marten-terminal-bench-harbor-laude-institute" ] }, { "slug": "lease-end", "name": "Lease End", "url": "https://ai.engineer/orgs/lease-end", "summary": "At AI Engineer World's Fair 2026, Lease End's Dan Bjornn described how a fine-tuned customer-intent model delivered reported revenue gains while accumulating labeling, retraining, and classification problems. His talk follows the team's rebuild around a model-agnostic agentic framework.", "speakerSlugs": [ "dan-bjornn" ], "talkSlugs": [ "your-fine-tuned-model-is-tech-debt-a-50x-roi-house-of-cards" ] }, { "slug": "legora", "name": "Legora", "url": "https://ai.engineer/orgs/legora", "summary": "At AI Engineer Europe 2026, Jacob Lauritzen argued that legal agents need collaboration interfaces richer than linear chat. Using contract review and litigation strategy, his talk examined how interface design can give people more control over complex automated work.", "speakerSlugs": [ "jacob-lauritzen" ], "talkSlugs": [ "agents-need-more-than-a-chat" ] }, { "slug": "leibniz-labs", "name": "Leibniz Labs", "url": "https://ai.engineer/orgs/leibniz-labs", "summary": "At AI Engineer World's Fair 2026, Erik Meijer of Leibniz Labs examined the risks of tool-enabled agents, focusing on prompt injection and unchecked side effects. His talk argues for enforceable safety guarantees before models can act.", "speakerSlugs": [ "erik-meijer" ], "talkSlugs": [ "i-ve-never-seen-anything-scarier-than-an-llm-with-tool-calls-erik-meijer-aka-headinthebox" ] }, { "slug": "lemonslice", "name": "LemonSlice", "url": "https://ai.engineer/orgs/lemonslice", "summary": "At AI Engineer World's Fair 2026, Sidney Primas presented LemonSlice's real-time visual agents through demonstrations of an interactive historical avatar and single-image characters. The talk considered emotional expression, evaluation, and future operating costs in pursuit of an avatar Turing test.", "speakerSlugs": [ "sidney-primas" ], "talkSlugs": [ "voice-agents-with-realtime-video-sidney-primas-lemonslice" ] }, { "slug": "lenses-io", "name": "Lenses.io", "url": "https://ai.engineer/orgs/lenses-io", "summary": "At AI Engineer Europe 2026, Tun Shwe and Jeremy Frenay presented MCP security through agent-oriented interface design, covering tool exposure, input validation, description poisoning, and sensitive context leaks. Their discussion grounded these risks in OWASP guidance and Lenses.io’s Kafka-focused open-source MCP server.", "speakerSlugs": [ "tun-shwe", "jeremy-frenay" ], "talkSlugs": [ "your-insecure-mcp-server-won-t-survive-production-tun-shwe-lenses" ] }, { "slug": "lexica", "name": "Lexica", "url": "https://ai.engineer/orgs/lexica", "summary": "At AI Engineer World's Fair 2025, Lexica’s Sharif Shameem presented curiosity as a way to discover AI capabilities before practical uses become clear. His talk revisits GPT-3 interface experiments and web agents, alongside Gemini 2.5 Pro video analysis.", "speakerSlugs": [ "sharif-shameem" ], "talkSlugs": [ "on-curiosity-sharif-shameem-lexica" ] }, { "slug": "lexisnexis", "name": "LexisNexis", "url": "https://ai.engineer/orgs/lexisnexis", "summary": "At AI Engineer World's Fair 2026, Sachin Kumar, affiliated with LexisNexis, presented independent research on sleeper-agent backdoors that behavioral evaluations can miss. His method trains a sparse autoencoder on differences between base-model and fine-tuned-model activations to identify signals of compromised behavior.", "speakerSlugs": [ "sachin-kumar" ], "talkSlugs": [ "your-llm-deception-monitor-is-broken-the-fix-is-in-the-training-data-sachin-kumar-lexisnexis" ] }, { "slug": "lindy", "name": "Lindy", "url": "https://ai.engineer/orgs/lindy", "summary": "At AI Engineer Summit 2023, Lindy’s Flo Crivello contrasted generative content tools with assistants that perform administrative work, including a meeting coordination demonstration. He argued that agentic AI could make business creation more accessible to individuals and small teams.", "speakerSlugs": [ "flo-crivello" ], "talkSlugs": [ "the-age-of-the-agent" ] }, { "slug": "livekit", "name": "Livekit", "url": "https://ai.engineer/orgs/livekit", "summary": "At AI Engineer World's Fair 2025, Tom Shapland, PhD examined why voice agents interrupt users when cascaded pipelines treat silence as a completed conversational turn. His talk connected human turn-taking with full-duplex approaches and a demonstration of semantic turn detection.", "speakerSlugs": [ "tom-shapland-phd" ], "talkSlugs": [ "why-chatgpt-keeps-interrupting-you" ] }, { "slug": "locally-ai", "name": "Locally AI", "url": "https://ai.engineer/orgs/locally-ai", "summary": "At AI Engineer Europe 2026, Adrien Grondin demonstrates running Gemma 4 on an iPhone with Apple’s MLX framework. His Locally AI appearance includes a claimed 40-tokens-per-second demo and a look at the broader MLX ecosystem for vision, audio, and video.", "speakerSlugs": [ "adrien-grondin" ], "talkSlugs": [ "running-gemma-4-on-device-40-tokens-s-on-iphone-with-mlx" ] }, { "slug": "los-alamos-national-laboratory", "name": "Los Alamos National Laboratory", "url": "https://ai.engineer/orgs/los-alamos-national-laboratory", "summary": "At AI Engineer World's Fair 2025, Mark Myshatyn discussed agentic AI for national-security science at Los Alamos National Laboratory, including inertial confinement fusion design and simulations on high-performance computing infrastructure. His talk connected these scientific applications with the practical constraints of deploying AI in a federal environment.", "speakerSlugs": [ "mark-myshatyn" ], "talkSlugs": [ "government-agents-ai-agents-meet-tough-regulations-mark-myshatyn-los-alamos-national-laboratory" ] }, { "slug": "lovable", "name": "Lovable", "url": "https://ai.engineer/orgs/lovable", "summary": "At AI Engineer Europe 2026, Benjamin Verbeek described how Lovable improves its AI coding platform through successful resolutions of stalled user sessions and agents reporting platform friction to Slack. The talk connects these feedback loops to measuring project completion for nontechnical users.", "speakerSlugs": [ "benjamin-verbeek" ], "talkSlugs": [ "how-lovable-self-improves-every-hour" ] }, { "slug": "luma-ai", "name": "Luma AI", "url": "https://ai.engineer/orgs/luma-ai", "summary": "At AI Engineer World's Fair 2025, Keegan McCallum recounts scaling Luma AI's Dream Machine to one million users in four days. His talk covers rapid GPU capacity expansion, operational problems with Triton Inference Server, and PyTorch-level optimization for serving generative video.", "speakerSlugs": [ "keegan-mccallum" ], "talkSlugs": [ "dream-machine-scaling-to-1m-users-in-4-days-keegan-mccallum-luma-ai" ] }, { "slug": "luminal", "name": "Luminal", "url": "https://ai.engineer/orgs/luminal", "summary": "At AI Engineer World's Fair 2025, Joe Fioti presented Luminal's approach to representing machine-learning models with a small set of primitive tensor operations and static computation graphs. His talk covered direct compilation to CUDA, bounded transformer dynamism, and buffer reuse informed by whole-workload visibility.", "speakerSlugs": [ "joe-fioti" ], "talkSlugs": [ "luminal-search-based-deep-learning-compilers-joe-fioti" ] }, { "slug": "lux-capital", "name": "Lux Capital", "url": "https://ai.engineer/orgs/lux-capital", "summary": "At AI Engineer Summit 2025, Grace Isford surveyed how stronger reasoning models, falling inference costs, open-model competition, and infrastructure investment create opportunities for AI agents. Her talk connected those developments with practical failures in handling user preferences, context, and execution.", "speakerSlugs": [ "grace-isford" ], "talkSlugs": [ "beyond-the-consensus-navigating-ai-s-frontier-in-2025" ] }, { "slug": "lyft", "name": "Lyft", "url": "https://ai.engineer/orgs/lyft", "summary": "At AI Engineer World's Fair 2026, Lyft speakers Nick Ung and Akshay Sharma discuss evaluating multi-turn customer-support agents before deployment. Their approach combines production-representative simulated users, LLM-as-a-judge grading, and explicit launch gates, including τ-bench-inspired simulations of LangGraph agents.", "speakerSlugs": [ "nick-ung", "akshay-sharma" ], "talkSlugs": [ "build-evals-that-actually-matter-nick-ung-akshay-sharma-lyft" ] }, { "slug": "m87-labs", "name": "M87 Labs", "url": "https://ai.engineer/orgs/m87-labs", "summary": "At AI Engineer World's Fair 2024, M87 Labs’ Vikhyat Korrapati presented Moondream, a compact open-source vision-language model for image questions, captioning, localization, and counting. The talk connects its browser-testing origins with an architecture combining pretrained SigLIP and Phi-1.5 components.", "speakerSlugs": [ "vikhyat-korrapati" ], "talkSlugs": [ "moondream-how-does-a-tiny-vision-model-slap-so-hard" ] }, { "slug": "machinecraft", "name": "Machinecraft", "url": "https://ai.engineer/orgs/machinecraft", "summary": "At AI Engineer World's Fair 2026, Rushabh Doshi described turning Machinecraft’s closely held manufacturing knowledge into Ira, a system of bounded specialist agents coordinated by Athena. His talk covers sales and customer qualification, combining vector, graph, and CRM data, and a forkable BrainOS architecture.", "speakerSlugs": [ "rushabh-doshi" ], "talkSlugs": [ "the-factory-that-dreams-39-ai-agents-no-framework" ] }, { "slug": "manufact-inc", "name": "Manufact, Inc", "url": "https://ai.engineer/orgs/manufact-inc", "summary": "At AI Engineer World's Fair 2026, Pietro Zullo presented MCP Apps as a way to add interactive, host-rendered interfaces to Model Context Protocol servers. His talk introduced Manufact’s mcp-use SDKs, Inspector, and cloud platform, alongside questions of application distribution and discovery.", "speakerSlugs": [ "pietro-zullo" ], "talkSlugs": [ "mcp-apps-primitives-discovery-and-the-future-of-software" ] }, { "slug": "manufactured", "name": "Manufactured", "url": "https://ai.engineer/orgs/manufactured", "summary": "At AI Engineer Summit 2025, Manufactured's affiliated speaker Kam Lasater builds a minimal AI agent without an agent framework. The session starts with an OpenAI LLM call and a strict JSON LLM-as-judge, then adds search, tool execution, and iterative control flow.", "speakerSlugs": [ "kam-lasater" ], "talkSlugs": [ "let-s-build-an-agent-from-scratch-kam-lasater" ] }, { "slug": "manus", "name": "Manus", "url": "https://ai.engineer/orgs/manus", "summary": "At AI Engineer Code 2025, Ivan Leo demonstrates building research and workflow agents with Manus 1.5 and the Manus API. The workshop connects hands-on API work with applications across Slack, browsers, email, and generated web apps.", "speakerSlugs": [ "ivan-leo" ], "talkSlugs": [ "building-intelligent-research-agents-with-manus" ] }, { "slug": "mastercard", "name": "Mastercard", "url": "https://ai.engineer/orgs/mastercard", "summary": "At AI Engineer World's Fair 2024, Ahmed Menshawy discussed the shift from structured-data machine learning to enterprise LLM applications. His talk connected generative-AI fraud detection with practical deployment concerns, including inference infrastructure, GPU and memory demands, and technical debt beyond model code.", "speakerSlugs": [ "ahmed-menshawy" ], "talkSlugs": [ "navigating-challenges-and-technical-debt-in-llms-deployment" ] }, { "slug": "maven-clinic", "name": "Maven Clinic", "url": "https://ai.engineer/orgs/maven-clinic", "summary": "At AI Engineer World's Fair 2026, Dan Feng described Maven Clinic's approach to becoming an AI-native healthcare company through internal adoption, AI-enabled products, and cultural change. His talk introduced Maven Intelligence as a shared orchestration layer and described changes to engineering workflows.", "speakerSlugs": [ "dan-feng" ], "talkSlugs": [ "how-to-build-an-ai-native-health-company" ] }, { "slug": "mckinsey-company", "name": "McKinsey & Company", "url": "https://ai.engineer/orgs/mckinsey-company", "summary": "At AI Engineer Code 2025, Martin Harrysson and Natasha Maniar argued that AI-assisted software development calls for redesigning operating models beyond adding coding tools to Agile teams. Their presentation connects changes in engineering workflows with upskilling, incentives, and organizational change.", "speakerSlugs": [ "martin-harrysson", "natasha-maniar" ], "talkSlugs": [ "moving-away-from-agile-what-s-next" ] }, { "slug": "mcp-apps", "name": "MCP Apps", "url": "https://ai.engineer/orgs/mcp-apps", "summary": "At AI Engineer World's Fair 2026, Ido Salomon and Liad Yosef explain how MCP-UI evolved into MCP Apps, an open Model Context Protocol extension. Their talk describes how services deliver branded, interactive application interfaces inside AI assistants instead of text-only responses.", "speakerSlugs": [ "ido-salomon", "liad-yosef" ], "talkSlugs": [ "mcp-apps-extending-the-frontier-ido-salomon-liad-yosef" ] }, { "slug": "meta-pytorch", "name": "Meta PyTorch", "url": "https://ai.engineer/orgs/meta-pytorch", "summary": "At AI Engineer Summit 2025, Soumith Chintala examined personal AI agents that keep sensitive context under users’ control. His talk connects unreliable answers to fragmented access across services and considers the privacy and delegated-action risks of assistants that act on a user’s behalf.", "speakerSlugs": [ "soumith-chintala" ], "talkSlugs": [ "what-does-it-take-to-build-a-personal-local-private-ai-agent-that-augments-you-deeply" ] }, { "slug": "method-financial", "name": "Method Financial", "url": "https://ai.engineer/orgs/method-financial", "summary": "At AI Engineer Summit 2025, Method Financial's Mustafa Ali joined OpenPipe's Kyle Corbitt to discuss scaling agents that process unstructured liability data. Their session describes an initial GPT-4 workflow whose production costs and prompt-engineering limitations prompted discussion of fine-tuned open-source models.", "speakerSlugs": [ "mustafa-ali" ], "talkSlugs": [ "how-we-scaled-500m-ai-agents-in-production-with-2-engineers" ] }, { "slug": "midjourney", "name": "Midjourney", "url": "https://ai.engineer/orgs/midjourney", "summary": "At AI Engineer World's Fair 2024, Midjourney-affiliated speaker Cheng Lou examined AI’s indirect effects on computing, learning, creativity, and communication. His talk, Second Order Effects, connects examples from GPU shader art and strategy games to possibilities for expanding human skills.", "speakerSlugs": [ "cheng-lou" ], "talkSlugs": [ "second-order-effects" ] }, { "slug": "mindmakers", "name": "Mindmakers", "url": "https://ai.engineer/orgs/mindmakers", "summary": "At AI Engineer World's Fair 2026, Alejandro Vidal examined model evaluation through item response theory. Using Epoch AI benchmark data, his talk argued that counting correct answers obscures question difficulty, uncertainty, and behavioral differences between similarly scoring models.", "speakerSlugs": [ "alejandro-vidal" ], "talkSlugs": [ "stop-evaluating-models-like-it-s-the-50s-alejandro-vidal-mindmakers" ] }, { "slug": "mit-media-lab", "name": "MIT Media Lab", "url": "https://ai.engineer/orgs/mit-media-lab", "summary": "At AI Engineer World's Fair 2026, MIT Media Lab’s Ramesh Raskar joins Maria Gorskikh in a discussion of open infrastructure for an agentic web. The session connects agent discovery, trustworthy identities, transactions, and coordination across organizational boundaries with a collaborative bazaar model.", "speakerSlugs": [ "ramesh-raskar" ], "talkSlugs": [ "the-agentic-web-and-the-bazaar-era-of-ai-ramesh-raskar-mit-media-lab" ] }, { "slug": "mixedbread", "name": "Mixedbread", "url": "https://ai.engineer/orgs/mixedbread", "summary": "At AI Engineer World's Fair 2026, Hanna Lichtenberg and Aamir Shakir examine how weak retrieval limits capable models on research tasks. Their Mixedbread presentation compares keyword-oriented search with semantic and late-interaction retrieval using BrowseComp+ and OfficeQA Pro, and reports preliminary benchmark results.", "speakerSlugs": [ "hanna-lichtenberg", "aamir-shakir" ], "talkSlugs": [ "how-we-taught-agents-to-use-good-retrieval-hanna-lichtenberg-mixedbread-ai" ] }, { "slug": "modular", "name": "Modular", "url": "https://ai.engineer/orgs/modular", "summary": "At AI Engineer World's Fair 2024, Chris Lattner presented MAX and Mojo as a unified, Python-oriented approach to model deployment and CPU/GPU programming. His talk framed developer productivity and hardware performance as connected goals for production AI infrastructure.", "speakerSlugs": [ "chris-lattner" ], "talkSlugs": [ "unlocking-developer-productivity-across-cpu-and-gpu-with-max" ] }, { "slug": "monday-com", "name": "monday.com", "url": "https://ai.engineer/orgs/monday-com", "summary": "At AI Engineer World's Fair 2026, Omri Bruchim and Tomer Ast explain why retrieving disconnected workplace records is insufficient for prioritizing work. Their proposed Monday world model precomputes relationships and user context for monday sidekick, combining durable behavioral patterns with current commitments.", "speakerSlugs": [ "omri-bruchim", "tomer-ast" ], "talkSlugs": [ "from-systems-of-record-to-systems-of-context" ] }, { "slug": "mongodb-voyage-ai", "name": "MongoDB / Voyage AI", "url": "https://ai.engineer/orgs/mongodb-voyage-ai", "summary": "At AI Engineer World's Fair 2025, Tengyu Ma examined retrieval-augmented generation for proprietary enterprise knowledge. His talk explains why he favors RAG over fine-tuning and long-context approaches, then surveys techniques for improving retrieval accuracy.", "speakerSlugs": [ "tengyu-ma" ], "talkSlugs": [ "rag-in-2025-state-of-the-art-and-the-road-forward" ] }, { "slug": "morph-labs", "name": "Morph Labs", "url": "https://ai.engineer/orgs/morph-labs", "summary": "At AI Engineer World's Fair 2025, Jesse Han presented Morph Labs’ vision for agent-native cloud infrastructure. His talk described Infinibranch’s virtualization, storage, and networking stack as a foundation for reversible execution, snapshots, and parallel reasoning, connecting those capabilities to agent swarms and verified reasoning.", "speakerSlugs": [ "jesse-han" ], "talkSlugs": [ "the-infrastructure-for-the-singularity" ] }, { "slug": "mozilla", "name": "Mozilla", "url": "https://ai.engineer/orgs/mozilla", "summary": "At AI Engineer World's Fair 2024, Stephen Hood and Justine Tunney presented Mozilla-backed llamafile, which packages large language models as portable, single-file executables for local use. Their presentation covered offline operation and included a summarization demonstration.", "speakerSlugs": [ "stephen-hood", "justine-tunney" ], "talkSlugs": [ "llamafile-bringing-ai-to-the-masses-with-fast-cpu-inference" ] }, { "slug": "mozilla-ai", "name": "Mozilla.ai", "url": "https://ai.engineer/orgs/mozilla-ai", "summary": "John Dickerson's appearance at AI Engineer World's Fair 2025 examines why evaluation and monitoring become commercially urgent as generative AI and autonomous agents enter production. The talk connects business KPIs, security, and quantitative risk with the shift toward agentic and multi-agent monitoring.", "speakerSlugs": [ "john-dickerson" ], "talkSlugs": [ "2025-is-the-year-of-evals-just-like-2024-and-2023-and" ] }, { "slug": "multinear", "name": "Multinear", "url": "https://ai.engineer/orgs/multinear", "summary": "At AI Engineer World's Fair 2025, Dmitry Kuchin presented an approach to improving generative AI reliability by deriving evaluations from user scenarios and business outcomes. His talk introduced the open-source Multinear evaluation platform and demonstrated using o3 to generate questions, checklists, and persona variations.", "speakerSlugs": [ "dmitry-kuchin" ], "talkSlugs": [ "practical-tactics-to-build-reliable-ai-apps-dmitry-kuchin-multinear" ] }, { "slug": "muna", "name": "Muna", "url": "https://ai.engineer/orgs/muna", "summary": "At AI Engineer Code 2025, Yusuf Olokoba of Muna presented a compiler for turning Python AI workloads into portable native binaries. His talk connects deployment across cloud, edge, and devices with access to open-source models through an OpenAI-compatible client.", "speakerSlugs": [ "yusuf-olokoba" ], "talkSlugs": [ "compilers-in-the-age-of-llms" ] }, { "slug": "mutagent", "name": "Mutagent", "url": "https://ai.engineer/orgs/mutagent", "summary": "Benedikt Sanftl and Burak Cemil Özafşar present an agentic engineering workflow at AI Engineer World's Fair 2026, connecting offline building and evaluation with online monitoring, trace diagnosis, and optimization. Their session also covers orchestration across development and observability tools, followed by a product demonstration.", "speakerSlugs": [ "benedikt-sanftl", "burak-cemil-ozafsar" ], "talkSlugs": [ "the-agentic-ai-engineer" ] }, { "slug": "n8n", "name": "n8n", "url": "https://ai.engineer/orgs/n8n", "summary": "At AI Engineer Europe 2026, Liam McGarrigle demonstrated building a visual AI agent for Gmail and Google Calendar management with n8n. The hands-on workshop paired conditional integrations with execution logs and human approval, connecting everyday automation tasks to practical oversight.", "speakerSlugs": [ "liam-mcgarrigle" ], "talkSlugs": [ "building-your-own-secure-ai-workflows-human-in-the-loop-automation-with-n8n" ] }, { "slug": "namespace", "name": "Namespace", "url": "https://ai.engineer/orgs/namespace", "summary": "At AI Engineer Europe 2026, Namespace-affiliated Hugo Santos joined Madison Faulkner to discuss how simultaneous agent-generated changes strain CI/CD pipelines and pull-request review. Their discussion proposes continuous compute and faster build-and-test infrastructure to support development at agent scale.", "speakerSlugs": [ "hugo-santos" ], "talkSlugs": [ "ci-cd-is-dead-agents-need-continuous-compute-and-computers-hugo-santos-and-madison-faulkner" ] }, { "slug": "nearform", "name": "Nearform", "url": "https://ai.engineer/orgs/nearform", "summary": "At AI Engineer World's Fair 2026, Nearform's Alfonso Graziano demonstrates how a coding agent can iteratively build and improve a production AI agent. The talk connects spec-driven engineering, golden datasets, and evaluation scorers with human oversight, using Mastra and an autoresearch-inspired AutoAgent loop.", "speakerSlugs": [ "alfonso-graziano" ], "talkSlugs": [ "agents-building-agents" ] }, { "slug": "nebius", "name": "Nebius", "url": "https://ai.engineer/orgs/nebius", "summary": "At AI Engineer Europe 2026, Ibragim Badertdinov presented SWE-rebench's continuous evaluation of coding agents on fresh GitHub software-engineering tasks. The talk covers containerized execution, regression testing, and task validation, alongside practical evaluation concerns such as harness differences, token costs, caching, and configuration drift.", "speakerSlugs": [ "ibragim-badertdinov" ], "talkSlugs": [ "swe-rebench-lessons-from-evaluating-coding-agents-on-real-software-engineering-tasks-ibragim-bad" ] }, { "slug": "neocognition", "name": "NeoCognition", "url": "https://ai.engineer/orgs/neocognition", "summary": "At AI Engineer World's Fair 2026, Yu Su distinguishes general intelligence from situated expertise. His talk connects agents' brittleness, inefficient token use, and computer-use difficulties to missing domain knowledge, arguing that intelligence alone does not explain reliable performance.", "speakerSlugs": [ "yu-su" ], "talkSlugs": [ "intelligence-continual-learning-expertise" ] }, { "slug": "nereu", "name": "Nereu", "url": "https://ai.engineer/orgs/nereu", "summary": "At AI Engineer World's Fair 2026, Arturo Nunez demonstrates game creation with Nereu's Bibi assistant, translating natural-language requests into assets, movement, animation, and gameplay behavior. His talk presents a browser-based JavaScript implementation aimed at reducing repetitive work in conventional game engines.", "speakerSlugs": [ "arturo-nunez" ], "talkSlugs": [ "the-next-game-engine-won-t-have-a-manual" ] }, { "slug": "new-computer", "name": "New Computer", "url": "https://ai.engineer/orgs/new-computer", "summary": "At AI Engineer Summit 2023, Samantha Whitmore and Jason Yuan presented AI interfaces that sense context and respond beyond text-only conversation. Their New Computer session paired a live demonstration with speculative concepts for headset interaction, social interfaces, and a generative canvas grounded in physical surroundings.", "speakerSlugs": [ "samantha-whitmore", "jason-yuan" ], "talkSlugs": [ "the-intelligent-interface" ] }, { "slug": "new-enterprise-associates-nea", "name": "New Enterprise Associates (NEA)", "url": "https://ai.engineer/orgs/new-enterprise-associates-nea", "summary": "At AI Engineer Europe 2026, NEA-affiliated Madison Faulkner joined Hugo Santos to discuss development infrastructure for autonomous coding agents. Their joint session argues that simultaneous agent-generated changes strain CI/CD pipelines and pull-request review, motivating continuous compute and faster build-and-test infrastructure.", "speakerSlugs": [ "madison-faulkner" ], "talkSlugs": [ "ci-cd-is-dead-agents-need-continuous-compute-and-computers-hugo-santos-and-madison-faulkner" ] }, { "slug": "new-generation-new-gen", "name": "New Generation (New Gen)", "url": "https://ai.engineer/orgs/new-generation-new-gen", "summary": "At AI Engineer World's Fair 2025, Adam Behrens of New Generation (New Gen) examined how AI agents could turn static online storefronts into conversational commerce. His talk covered translating ambiguous customer intent into specific products, integrating merchant data, and generating shopping interfaces dynamically.", "speakerSlugs": [ "adam-behrens" ], "talkSlugs": [ "machines-of-buying-selling-grace" ] }, { "slug": "news-corp", "name": "News Corp", "url": "https://ai.engineer/orgs/news-corp", "summary": "At AI Engineer Summit 2025, News Corp–affiliated speaker Manish Sanwal presented Layered Chain-of-Thought for multi-agent AI systems. His talk examines how checking intermediate reasoning steps can make outputs easier to audit, reproduce, and correct.", "speakerSlugs": [ "manish-sanwal" ], "talkSlugs": [ "stop-guessing-build-robust-ai-with-layered-cot" ] }, { "slug": "nori", "name": "Nori", "url": "https://ai.engineer/orgs/nori", "summary": "At AI Engineer World's Fair 2026, Nori's Amol Kapoor presented HTML as a structured medium for agents to produce graphics that browsers render into slides, documents, videos, and PDFs. He also described using HTML and CSS for branded visual assets.", "speakerSlugs": [ "amol-kapoor" ], "talkSlugs": [ "html-is-all-you-need-for-agents-to-make-graphics" ] }, { "slug": "normal-computing", "name": "Normal Computing", "url": "https://ai.engineer/orgs/normal-computing", "summary": "At AI Engineer World's Fair 2024, Phoebe Klett presented Extended Mind Transformers: decoder-only language models augmented with cached external key-value memories. Her talk contrasted retrieval through model-native query/key similarity with conventional RAG and explained how the approach avoids fine-tuning.", "speakerSlugs": [ "phoebe-klett" ], "talkSlugs": [ "going-beyond-rag-extended-mind-transformers" ] }, { "slug": "northwestern-mutual", "name": "Northwestern Mutual", "url": "https://ai.engineer/orgs/northwestern-mutual", "summary": "At AI Engineer Code 2025, Asaf Bord describes developing Northwestern Mutual's GenBI business-intelligence agent through incremental projects that earn funding and trust. The talk connects enterprise adoption to a staged rollout, beginning with BI experts before reaching business managers.", "speakerSlugs": [ "asaf-bord" ], "talkSlugs": [ "small-bets-big-impact-building-genbi-at-a-fortune-100" ] }, { "slug": "notable-capital", "name": "Notable Capital", "url": "https://ai.engineer/orgs/notable-capital", "summary": "AI startup fundraising is the focus of Chelcie Taylor’s recorded appearance at AI Engineer World's Fair 2025. In a joint workshop with Dani Grant of Jam, the discussion covers investor psychology, founder outreach, differentiated AI pitches, ideal customer profiles, and team evaluation.", "speakerSlugs": [ "chelcie-taylor" ], "talkSlugs": [ "the-ai-engineer-s-guide-to-raising-vc-dani-grant-jam-chelcie-taylor-notable-capital" ] }, { "slug": "notius-labs", "name": "Notius Labs", "url": "https://ai.engineer/orgs/notius-labs", "summary": "At AI Engineer Europe 2026, Christopher Lovejoy discussed building vertical AI products through systematic use of domain expertise. His talk contrasts Granola's meeting-note workflow with Tandem's clinical-scribe use case and argues that incorporating expert judgment matters more than choosing the most sophisticated models.", "speakerSlugs": [ "christopher-lovejoy" ], "talkSlugs": [ "how-to-leverage-domain-expertise-chris-lovejoy-notius-labs" ] }, { "slug": "nx", "name": "Nx", "url": "https://ai.engineer/orgs/nx", "summary": "At AI Engineer World's Fair 2026, Victor Savkin examines how fragmented repository visibility and forgotten sessions constrain coding agents. He presents Nx-built Polygraph as an agent-agnostic meta-harness for preserving shared organizational context and helping agents work across repositories.", "speakerSlugs": [ "victor-savkin" ], "talkSlugs": [ "a-genius-with-amnesia" ] }, { "slug": "octoai", "name": "OctoAI", "url": "https://ai.engineer/orgs/octoai", "summary": "Thierry Moreau and Pedro Torruella's workshop at AI Engineer World's Fair 2024 examines fine-tuning smaller Llama 3 models for function calling, classification, and PII redaction. The session connects task specialization to deployment costs and GPU constraints when comparing smaller models with larger proprietary alternatives.", "speakerSlugs": [ "thierry-moreau", "pedro-torruella" ], "talkSlugs": [ "llm-quality-optimization-bootcamp" ] }, { "slug": "ogilvy", "name": "Ogilvy", "url": "https://ai.engineer/orgs/ogilvy", "summary": "At AI Engineer World's Fair 2026, Ogilvy-affiliated speaker Abed Matini demonstrates a local-first employee-handbook FAQ assistant. His talk covers document ingestion, chunking, and framework-light hybrid RAG using PostgreSQL retrieval and Reciprocal Rank Fusion.", "speakerSlugs": [ "abed-matini" ], "talkSlugs": [ "bypassing-the-multimodal-tax-framework-free-hybrid-rag-raw-sql-rrf-and-live-ui-telemetry" ] }, { "slug": "oliver", "name": "OLIVER", "url": "https://ai.engineer/orgs/oliver", "summary": "At AI Engineer Europe 2026, OLIVER-affiliated Angus J. McLean presented bounded autonomy as a practical approach to agentic advertising workflows. His talk connects creative constraints and simple systems with short real-world feedback loops, ending with social-post clustering for creative and strategic insights.", "speakerSlugs": [ "angus-j-mclean" ], "talkSlugs": [ "bounded-autonomy-between-free-will-and-determinism" ] }, { "slug": "ollama", "name": "Ollama", "url": "https://ai.engineer/orgs/ollama", "summary": "Compressing open models for local and edge deployment was the focus of a joint panel featuring Ollama-affiliated Parth Sareen at AI Engineer World's Fair 2026. Moderated by Chris Alexiuk, the discussion examined quantization and the tradeoffs between model size and quality.", "speakerSlugs": [ "parth-sareen" ], "talkSlugs": [ "compression-at-the-edge" ] }, { "slug": "omnara", "name": "Omnara", "url": "https://ai.engineer/orgs/omnara", "summary": "At AI Engineer World's Fair 2026, Ishaan Sehgal argued that an agent's durable identity lives in its session log. His talk connects persistent history to crash recovery, preserved permission prompts, branching, and portability across models, treating context and interfaces as projections of that history.", "speakerSlugs": [ "ishaan-sehgal" ], "talkSlugs": [ "the-log-is-the-agent" ] }, { "slug": "onlay", "name": "Onlay", "url": "https://ai.engineer/orgs/onlay", "summary": "At AI Engineer World's Fair 2026, Vasant Kearney of Onlay presents X12 healthcare transaction standards as a structured harness for agents handling claims, payer interactions, and payment workflows. The talk argues that standardized structures can constrain costly, error-prone multistep reasoning.", "speakerSlugs": [ "vasant-kearney" ], "talkSlugs": [ "healthcare-s-agent-bytecode-x12-as-the-harness-for-ai-agents" ] }, { "slug": "openaudio-fish-audio", "name": "OpenAudio / Fish Audio", "url": "https://ai.engineer/orgs/openaudio-fish-audio", "summary": "OpenAudio / Fish Audio's S1 voice model appears in the HF0 Residency startup showcase at AI Engineer World's Fair 2025, with Shijia Liao listed as its affiliated speaker. The shared session places voice generation alongside creative tools, enterprise data, and model routing.", "speakerSlugs": [ "shijia-liao" ], "talkSlugs": [ "the-next-unicorns-7-top-ai-startups-from-the-hf0-residency" ] }, { "slug": "openclaw", "name": "OpenClaw", "url": "https://ai.engineer/orgs/openclaw", "summary": "At AI Engineer Europe 2026, Radek Sienkiewicz described gradually giving a personal AI agent access to his communications, files, calendar, and operating system. His talk connects WhatsApp, Telegram, and Discord interfaces with an Obsidian knowledge base and QMD search.", "speakerSlugs": [ "radek-sienkiewicz" ], "talkSlugs": [ "i-gave-an-ai-agent-the-keys-to-my-life-here-s-what-happened" ] }, { "slug": "openclaw-textcortex-at-the-time-of-recording", "name": "OpenClaw; TextCortex at the time of recording", "url": "https://ai.engineer/orgs/openclaw-textcortex-at-the-time-of-recording", "summary": "At AI Engineer Europe 2026, Onur Solmaz discusses OpenClaw maintainer work and earlier TextCortex coding-agent development through acpx and the Agent Client Protocol. His demonstration connects parallel Codex workflows in Discord channels with a discussion of Kubernetes and scalable multi-agent provisioning.", "speakerSlugs": [ "onur-solmaz" ], "talkSlugs": [ "scaling-agents-on-kubernetes-with-acpx-and-acp" ] }, { "slug": "opencode", "name": "OpenCode", "url": "https://ai.engineer/orgs/opencode", "summary": "At AI Engineer Code 2025, OpenCode contributor Dax Raad examines product building through marketing, onboarding, and lasting depth. He argues that faster coding with AI leaves taste, creativity, deep understanding, and difficult execution essential to building a successful product.", "speakerSlugs": [ "dax-raad" ], "talkSlugs": [ "ai-changes-nothing-dax-raad-opencode" ] }, { "slug": "opengov", "name": "OpenGov", "url": "https://ai.engineer/orgs/opengov", "summary": "At AI Engineer World's Fair 2026, Gabe De Mesa described how OpenGov built and scaled OG Assist, an AI-agent experience embedded in government-software products. The talk covers production architecture, tool calling, and the team's use of Effect and TypeScript.", "speakerSlugs": [ "gabe-de-mesa" ], "talkSlugs": [ "agents-in-production-how-opengov-built-and-scaled-og-assist" ] }, { "slug": "openprose", "name": "OpenProse", "url": "https://ai.engineer/orgs/openprose", "summary": "At AI Engineer World's Fair 2026, Raymond Weitekamp of OpenProse argued that dependable coding agents need outcome-oriented orchestration beyond stronger models. His talk contrasted Claude Code dynamic workflows with OpenProse approaches to recursive agent execution, alongside discussion of long-context processing and memory.", "speakerSlugs": [ "raymond-weitekamp" ], "talkSlugs": [ "recursive-coding-agents" ] }, { "slug": "orbis-operations", "name": "Orbis Operations", "url": "https://ai.engineer/orgs/orbis-operations", "summary": "Luis Romero-Sevilla's appearance at AI Engineer World's Fair 2026 centers on Extended Cache Augmented Generation (ECAG) for rapidly changing, interconnected document collections. He presents parallel cached context buckets queried by a supervising model that synthesizes answers, contrasting the approach with vector-based RAG and GraphRAG.", "speakerSlugs": [ "luis-romero-sevilla" ], "talkSlugs": [ "when-all-context-matters-extended-cache-augmented-generation-ecag" ] }, { "slug": "orbital", "name": "Orbital", "url": "https://ai.engineer/orgs/orbital", "summary": "At AI Engineer World's Fair 2025, Andrew Thompson explained the engineering and domain-expert effort needed when frontier model upgrades break prompts or introduce regressions. His Orbital Copilot demonstration connected this maintenance burden to document processing, task decomposition, and lawyer-reviewed outputs for real-estate legal work.", "speakerSlugs": [ "andrew-thompson" ], "talkSlugs": [ "buy-now-maybe-pay-later-dealing-with-prompt-tax-while-staying-at-the-frontier-andrew-thompson" ] }, { "slug": "oxylabs", "name": "Oxylabs", "url": "https://ai.engineer/orgs/oxylabs", "summary": "At AI Engineer World's Fair 2026, Patricija Žemaitytė discussed infrastructure for keeping AI supplied with fresh public web data. Her talk covered the evolution of a multimodal video API, faster search retrieval, and proxy-backed scaling under production demands.", "speakerSlugs": [ "patricija-zemaityte" ], "talkSlugs": [ "how-web-data-infrastructure-powers-the-next-generation-of-ai" ] }, { "slug": "palo-alto-networks", "name": "Palo Alto Networks", "url": "https://ai.engineer/orgs/palo-alto-networks", "summary": "At AI Engineer World's Fair 2024, Gunjan Patel of Palo Alto Networks presented a personal, noncommercial approach to moving time-consuming AI-assisted development into CI. His asynchronous ghostpilot concept covers code cleanup and context-aware test generation, with proposed fixes subject to human review.", "speakerSlugs": [ "gunjan-patel" ], "talkSlugs": [ "self-evolving-code-with-ai-enhancing-quality-and-security-in-ci" ] }, { "slug": "patho-ai", "name": "Patho.ai", "url": "https://ai.engineer/orgs/patho-ai", "summary": "At AI Engineer World's Fair 2025, Patho.ai's Chin Keong Lam presented knowledge-augmented generation as a graph-structured alternative to vector-only RAG for expert advisory systems. His talk connected knowledge, experience, insight, and feedback in a decision framework, illustrated through a competitive-analysis chatbot.", "speakerSlugs": [ "chin-keong-lam" ], "talkSlugs": [ "wisdom-driven-knowledge-augmented-generation-at-scale" ] }, { "slug": "perpetual", "name": "Perpetual", "url": "https://ai.engineer/orgs/perpetual", "summary": "At AI Engineer World's Fair 2024, Perpetual's Benjamin Stein discussed how recognizable personalities, voices, forms, and workplace roles make AI agents easier to understand. His talk examines the accompanying rise in customer expectations and gender-related design biases.", "speakerSlugs": [ "benjamin-stein" ], "talkSlugs": [ "personality-driven-development-exploring-the-frontier-of-agents-with-attitude" ] }, { "slug": "pff", "name": "PFF", "url": "https://ai.engineer/orgs/pff", "summary": "At AI Engineer Europe 2026, PFF's Mike Spitz presented an engineering case study focused on making AI agents faster. His talk describes a workflow connecting specifications, lightweight design documents, generated tickets, and pull requests, with automated coordination replacing standups and sprint refinement.", "speakerSlugs": [ "mike-spitz" ], "talkSlugs": [ "agents-don-t-do-standups-building-the-post-engineer-engineering-org" ] }, { "slug": "pfizer", "name": "Pfizer", "url": "https://ai.engineer/orgs/pfizer", "summary": "At AI Engineer Summit 2025, Pfizer’s Jonathan Lowe joined Neo4j’s Stephen Chin to discuss grounding enterprise generative AI with knowledge graphs and GraphRAG. The discussion connected life-sciences manufacturing and knowledge-transfer challenges with graph-based document chunking and contextual retrieval.", "speakerSlugs": [ "jonathan-lowe" ], "talkSlugs": [ "anchoring-enterprise-genai-with-knowledge-graphs" ] }, { "slug": "phaidra", "name": "Phaidra", "url": "https://ai.engineer/orgs/phaidra", "summary": "At AI Engineer World's Fair 2026, Phaidra's Raahul Singh and Vanč Levstik examined why LLMs struggle with large industrial equipment inventories. Their talk connects nearly identical sensor names, finite context windows, and unreliable retrieval to failures in identifying equipment reliably.", "speakerSlugs": [ "raahul-singh", "vanc-levstik" ], "talkSlugs": [ "semantic-blindness-500-000-sensors-confused-an-llm-raahul-singh-vanc-levstik-phaidra" ] }, { "slug": "philo-ventures", "name": "Philo Ventures", "url": "https://ai.engineer/orgs/philo-ventures", "summary": "Todd Fisher of Philo Ventures demonstrates an AI-assisted speaking guitar at AI Engineer World's Fair 2026, combining pitch detection, synthesized notes, and generated speech. His presentation introduces JUCE and digital audio workstation plug-ins while encouraging creative personal projects with AI.", "speakerSlugs": [ "todd-fisher" ], "talkSlugs": [ "while-my-guitar-gently-speaks" ] }, { "slug": "physical-intelligence", "name": "Physical Intelligence", "url": "https://ai.engineer/orgs/physical-intelligence", "summary": "At AI Engineer World's Fair 2025, Quan Vuong and Jost Tobias Springenberg discuss how advances in general AI are accelerating general-purpose robotics. Their talk connects pretrained vision-language models with robot control, showing household cleaning in previously unseen environments and end-to-end coffee preparation while acknowledging unresolved challenges.", "speakerSlugs": [ "quan-vuong", "jost-tobias-springenberg" ], "talkSlugs": [ "robotics-why-now" ] }, { "slug": "pi", "name": "PI", "url": "https://ai.engineer/orgs/pi", "summary": "At AI Engineer Europe 2026, Mario Zechner presented pi as a minimal, extensible coding-agent harness built around direct control of model context. His talk connects dissatisfaction with opaque agent behavior to concrete choices in pi's provider abstraction, agent loop, terminal UI, and extensions.", "speakerSlugs": [ "mario-zechner" ], "talkSlugs": [ "pi-coding-agent" ] }, { "slug": "pieces", "name": "Pieces", "url": "https://ai.engineer/orgs/pieces", "summary": "Offline developer tooling connects Pieces to Emma Ning’s Foundry Local presentation at AI Engineer World's Fair 2025. Tsavo Knott is the credited Pieces participant in a session that includes an embedded customer testimonial alongside Microsoft’s presentation of on-device AI.", "speakerSlugs": [ "tsavo-knott" ], "talkSlugs": [ "foundry-local-cutting-edge-ai-experiences-on-device-with-onnx-runtime-and-olive-emma-ning-micros" ] }, { "slug": "ping-labs", "name": "Ping Labs", "url": "https://ai.engineer/orgs/ping-labs", "summary": "At AI Engineer World's Fair 2026, Ping Labs' affiliated speaker Theo Browne argues that increasingly capable AI coding models change what small software teams can attempt. His talk connects changing development capabilities to broader product scope and a reassessment of inherited engineering constraints.", "speakerSlugs": [ "theo-browne" ], "talkSlugs": [ "everything-we-knew-about-software-has-changed-theo-browne" ] }, { "slug": "polar-signals", "name": "Polar Signals", "url": "https://ai.engineer/orgs/polar-signals", "summary": "At AI Engineer World's Fair 2025, Matthias Loibl of Polar Signals presented continuous GPU profiling that combines NVIDIA NVML metrics with low-overhead Linux eBPF CPU profiles. His demonstration connected GPU underutilization to Python and CUDA call stacks in flame charts.", "speakerSlugs": [ "matthias-loibl" ], "talkSlugs": [ "maximize-gpu-efficiency-with-continuous-profiling-for-gpus" ] }, { "slug": "pomerium", "name": "Pomerium", "url": "https://ai.engineer/orgs/pomerium", "summary": "At AI Engineer Europe 2026, Pomerium-affiliated Nick Taylor presented trusted-proxy authentication for OpenClaw, protecting its gateway behind an identity-aware proxy while avoiding repeated WebSocket token entry. He also demonstrated building an MCP application using conference speaker data from an OpenClaw workspace.", "speakerSlugs": [ "nick-taylor" ], "talkSlugs": [ "claws-out-securing-and-building-with-openclaw" ] }, { "slug": "portia-ai", "name": "Portia AI", "url": "https://ai.engineer/orgs/portia-ai", "summary": "At AI Engineer World's Fair 2025, Mounir Mouawad reflected on building an open-source SDK for production AI agents in regulated industries. His talk contrasted established-company product development with discovering emerging customer needs and earning distribution through partnerships, integration documentation, and developer communities.", "speakerSlugs": [ "mounir-mouawad" ], "talkSlugs": [ "from-pm-at-stripe-to-building-an-ai-startup-a-recent-founder-s-journey-mounir-mouawad" ] }, { "slug": "position2", "name": "Position²", "url": "https://ai.engineer/orgs/position2", "summary": "At AI Engineer World's Fair 2026, Dr. Sajjan Kanukolanu discussed B2B go-to-market agents that combine CRM records, conversations, and activity signals to prioritize buyers. His talk frames this work as coordinated AI, integration, and architecture beyond adding a chatbot to existing workflows.", "speakerSlugs": [ "dr-sajjan-kanukolanu" ], "talkSlugs": [ "build-the-ai-gtm-agent-that-knows-the-buyer-before-the-first-message" ] }, { "slug": "postman", "name": "Postman", "url": "https://ai.engineer/orgs/postman", "summary": "At AI Engineer Europe 2026, Postman's Ruben Casas examined how agent interfaces can progress from predefined components to dynamically generated UI. His talk used a weather-agent experiment to illustrate how MCP Apps can embed interactive third-party experiences in AI clients.", "speakerSlugs": [ "ruben-casas" ], "talkSlugs": [ "generative-ui-for-mcp" ] }, { "slug": "prediction-guard", "name": "Prediction Guard", "url": "https://ai.engineer/orgs/prediction-guard", "summary": "At AI Engineer World's Fair 2024, Daniel Whitenack presented a practical checklist for safer enterprise LLM deployment. His talk addressed hallucination risks, insecure open-model dependencies, sensitive-data exposure, and prompt injection, grounding Prediction Guard's recorded appearance in application security and privacy concerns.", "speakerSlugs": [ "daniel-whitenack" ], "talkSlugs": [ "llm-safeguards-security-privacy-compliance-anti-hallucination" ] }, { "slug": "privacera", "name": "Privacera", "url": "https://ai.engineer/orgs/privacera", "summary": "At AI Engineer Summit 2025, Don Bosco Durai addressed enterprise AI agent security, examining risks from shared-process architectures, service credentials, insecure LLMs, and regulated data access. His talk also introduced the open-source PAIG.ai security and safety project.", "speakerSlugs": [ "don-bosco-durai" ], "talkSlugs": [ "balancing-innovation-with-security-safety" ] }, { "slug": "programma-labs", "name": "Programma Labs", "url": "https://ai.engineer/orgs/programma-labs", "summary": "At AI Engineer World's Fair 2026, Pierluca D'Oro examined how deterministic computer-use benchmarks can reward blind replay scripts as much as frontier agents. His talk introduced PRISM principles and the DigiWorld Android benchmark as approaches to building more trustworthy evaluation environments.", "speakerSlugs": [ "pierluca-d-oro" ], "talkSlugs": [ "computer-use-at-the-edge-of-the-statistical-precipice" ] }, { "slug": "progress-software", "name": "Progress Software", "url": "https://ai.engineer/orgs/progress-software", "summary": "At AI Engineer World's Fair 2026, Progress Software's affiliated speaker Kathryn Grayson Nanz discussed designing AI applications around real user problems. Drawing on Macintosh interface evolution, she recommended introducing unfamiliar capabilities gradually without assuming users understand prompts, hallucinations, or retrieval augmented generation.", "speakerSlugs": [ "kathryn-grayson-nanz" ], "talkSlugs": [ "the-ux-of-ai-making-ai-powered-apps-your-users-don-t-hate" ] }, { "slug": "project-nanda", "name": "Project NANDA", "url": "https://ai.engineer/orgs/project-nanda", "summary": "At AI Engineer World's Fair 2026, Project NANDA's Maria Gorskikh joins Ramesh Raskar in a discussion of open infrastructure for agents to discover one another, establish trustworthy identities, transact, and coordinate across organizational boundaries.", "speakerSlugs": [ "maria-gorskikh" ], "talkSlugs": [ "the-agentic-web-and-the-bazaar-era-of-ai-ramesh-raskar-mit-media-lab" ] }, { "slug": "promptlayer", "name": "PromptLayer", "url": "https://ai.engineer/orgs/promptlayer", "summary": "At AI Engineer Code 2025, Jared Zoneraich examined Claude Code and other coding agents through simple prompt-driven architectures, isolated-context sub-agents, and tool calls. His workshop contrasted prompting with DAG-based orchestration and compared Claude Code with Codex.", "speakerSlugs": [ "jared-zoneraich" ], "talkSlugs": [ "how-claude-code-works" ] }, { "slug": "promptql", "name": "PromptQL", "url": "https://ai.engineer/orgs/promptql", "summary": "At AI Engineer World's Fair 2025, Anushrut Gupta presented PromptQL's approach to working with changing enterprise data. He argued that waiting for perfectly prepared data is impractical when schemas, business definitions, and organizational knowledge keep changing.", "speakerSlugs": [ "anushrut-gupta" ], "talkSlugs": [ "data-readiness-is-a-myth-reliable-ai-with-an-agentic-semantic-layer-anushrut-gupta-promptql" ] }, { "slug": "prosodica", "name": "Prosodica", "url": "https://ai.engineer/orgs/prosodica", "summary": "At AI Engineer World's Fair 2026, Prosodica’s Sohail Shaikh and Ankush Rastogi presented semantic routing for growing agent tool catalogs. Their talk describes using embeddings, vector search, and just-in-time context injection to give each request relevant capabilities without loading every tool schema into every prompt.", "speakerSlugs": [ "sohail-shaikh", "ankush-rastogi" ], "talkSlugs": [ "the-100-tool-agent-is-a-trap-scaling-with-semantic-routers-and-jit-context" ] }, { "slug": "pruna-ai", "name": "Pruna AI", "url": "https://ai.engineer/orgs/pruna-ai", "summary": "At AI Engineer Europe 2026, Bertrand Charpentier examined why image model rankings change across benchmarks and use cases. His talk challenged reliance on a single quality metric, highlighting limits of human inspection and CLIP scores alongside the compute and energy costs of evaluation.", "speakerSlugs": [ "bertrand-charpentier" ], "talkSlugs": [ "20-days-of-compute-vs-7-hours-rethinking-what-state-of-the-art-means-bertrand-charpentier-pruna" ] }, { "slug": "pulley", "name": "Pulley", "url": "https://ai.engineer/orgs/pulley", "summary": "At AI Engineer World's Fair 2025, Charlie Guo described using Pulley sales-call transcripts to sharpen customer profiles and extract customer intelligence in two weeks. The analysis became a reusable internal tool with search, filters, and exports.", "speakerSlugs": [ "charlie-guo" ], "talkSlugs": [ "analyzing-10-000-sales-calls-with-ai-in-2-weeks" ] }, { "slug": "pyannoteai", "name": "pyannoteAI", "url": "https://ai.engineer/orgs/pyannoteai", "summary": "At AI Engineer Europe 2026, Hervé Bredin presented speaker diarization as a way to recover conversational context that speech-to-text misses. The talk connected pyannote with Whisper and downstream language-model workflows, including meeting summaries and podcast analysis.", "speakerSlugs": [ "herve-bredin" ], "talkSlugs": [ "beyond-transcription-building-voice-ai-that-actually-understands-conversations" ] }, { "slug": "qdrant", "name": "Qdrant", "url": "https://ai.engineer/orgs/qdrant", "summary": "At AI Engineer World's Fair 2024, Qdrant-affiliated Atita Arora joined Deanna Emery of Quotient to discuss improving retrieval-augmented generation through systematic evaluation. Their session compares naive and advanced pipelines, examining query rewriting, re-ranking, context sizing, scalability, and security.", "speakerSlugs": [ "atita-arora" ], "talkSlugs": [ "navigating-rag-optimization-with-an-evaluation-driven-compass" ] }, { "slug": "quantyca", "name": "Quantyca", "url": "https://ai.engineer/orgs/quantyca", "summary": "At AI Engineer Europe 2026, Quantyca's Mauro Luchetti and Mattia Redaelli joined Amplifon's Sonny Merla to discuss enterprise AI governance through Amplifon's AmplifAI program. The session describes a centralized AI gateway and registries for MCP servers and A2A agents that address fragmented development.", "speakerSlugs": [ "mauro-luchetti", "mattia-redaelli" ], "talkSlugs": [ "one-registry-to-rule-them-all-sonny-merla-mauro-luchetti-mattia-redaelli-quantyca" ] }, { "slug": "radicait", "name": "RADiCAIT", "url": "https://ai.engineer/orgs/radicait", "summary": "At AI Engineer World's Fair 2026, Sina Shahandeh examined why autonomous research agents need explicit hypothesis generation beyond code and hyperparameter optimization. RADiCAIT's CT-to-synthetic-PET imaging work provided the case study for experiments involving GAN architectures, imaging representations, and evaluation.", "speakerSlugs": [ "sina-shahandeh" ], "talkSlugs": [ "autonomous-agents-for-scientific-tasks-sina-shahandeh-radicait" ] }, { "slug": "railway", "name": "Railway", "url": "https://ai.engineer/orgs/railway", "summary": "At AI Engineer Code 2025, Mahmoud Abdelwahab demonstrates railway-autofix, a prototype that monitors Railway deployments for memory leaks, HTTP errors, latency, and resource anomalies. It uses a coding agent to prepare corrective GitHub pull requests in response to detected problems.", "speakerSlugs": [ "mahmoud-abdelwahab" ], "talkSlugs": [ "infra-that-fixes-itself-thanks-to-coding-agents-mahmoud-abdelwahab-railway" ] }, { "slug": "rasgo", "name": "Rasgo", "url": "https://ai.engineer/orgs/rasgo", "summary": "At AI Engineer Summit 2025, Patrick Dougherty described rebuilding an enterprise-data product around autonomous AI agents. His talk covers goal-directed behavior, tool access, retrieval design, and model-specific interfaces, grounding Rasgo’s recorded presence in practical lessons from developing agents for enterprise analytics.", "speakerSlugs": [ "patrick-dougherty" ], "talkSlugs": [ "how-to-build-ai-agents-that-actually-work" ] }, { "slug": "ratel", "name": "Ratel", "url": "https://ai.engineer/orgs/ratel", "summary": "At AI Engineer World's Fair 2026, Ratel's Roberto Stagi presented a case for building production AI agents within a unified TypeScript application stack. His talk connects that choice to the shift from model research toward application and agent development.", "speakerSlugs": [ "roberto-stagi" ], "talkSlugs": [ "a-song-of-types-and-agents" ] }, { "slug": "re-skill", "name": "Re-skill", "url": "https://ai.engineer/orgs/re-skill", "summary": "At AI Engineer Summit 2025, Muhtesem presented a collaboration between Re-skill and Diffusion Studio on an open-source Python video-editing agent. The system uses Playwright to operate a browser editor, generates JavaScript or TypeScript compositions, and renders video through WebCodecs.", "speakerSlugs": [ "muhtesem" ], "talkSlugs": [ "this-video-was-edited-with-ai-agent-but-how" ] }, { "slug": "re-ma-ai", "name": "re:ma AI", "url": "https://ai.engineer/orgs/re-ma-ai", "summary": "At AI Engineer Summit 2023, Reid Mayo presented a practical curriculum for experienced full-stack developers entering AI engineering without prior machine-learning experience. His talk connects language-model fundamentals and prompt engineering with API practice, modular applications, and retrieval-augmented generation.", "speakerSlugs": [ "reid-mayo" ], "talkSlugs": [ "shift-left-how-to-become-an-ai-engineer-from-a-full-stack-background" ] }, { "slug": "reactor", "name": "Reactor", "url": "https://ai.engineer/orgs/reactor", "summary": "At AI Engineer World's Fair 2026, Reactor's Ahmed Ahres presents world models as real-time, interactive video. His talk contrasts continuous, controllable generation with conventional video models and considers applications in advertising, immersive education, audience-directed livestreaming, and video editing.", "speakerSlugs": [ "ahmed-ahres" ], "talkSlugs": [ "the-next-medium-why-real-time-interactive-video-changes-everything-ahmed-ahres-reactor" ] }, { "slug": "rechat", "name": "Rechat", "url": "https://ai.engineer/orgs/rechat", "summary": "At AI Engineer World's Fair 2024, Rechat's Emil Sedgh joined Hamel Husain to discuss domain-specific evaluations for Lucy, Rechat's real-estate AI assistant. Their session traces the application's development from an unreliable GPT-3.5 prototype toward production through repeated evaluation and prompt-engineering cycles.", "speakerSlugs": [ "emil-sedgh" ], "talkSlugs": [ "how-to-construct-domain-specific-llm-evaluation-systems" ] }, { "slug": "reelful", "name": "Reelful", "url": "https://ai.engineer/orgs/reelful", "summary": "At AI Engineer World's Fair 2026, Ekaterina Deyneka presented Reelful's approach to turning user photos, footage, and lightweight prompts into edited videos. The talk covers working with incomplete footage, animating photos, and using Remotion to compose the output.", "speakerSlugs": [ "ekaterina-deyneka" ], "talkSlugs": [ "building-an-agentic-video-editor-for-mass-consumer" ] }, { "slug": "ref", "name": "Ref.", "url": "https://ai.engineer/orgs/ref", "summary": "At AI Engineer World's Fair 2026, Matt Dailey described velocity sickness: faster individual engineering can overwhelm teams with pull requests, fragmented initiatives, and opaque automated decisions. His talk examines how teams can direct AI-driven speed toward coherent ideas and meaningful product impact.", "speakerSlugs": [ "matt-dailey" ], "talkSlugs": [ "velocity-sickness-what-happens-when-your-whole-team-gets-10x-faster" ] }, { "slug": "reforge", "name": "Reforge", "url": "https://ai.engineer/orgs/reforge", "summary": "At AI Engineer World's Fair 2025, Brian Balfour examined how rapid AI launches, incumbent platforms, and heavily funded startups make product differentiation harder. He introduced Reforge's AI research and project-management agents while discussing how to build products around unmet customer needs.", "speakerSlugs": [ "brian-balfour" ], "talkSlugs": [ "survive-the-ai-knife-fight-building-products-that-win" ] }, { "slug": "relai-university-of-maryland-college-park", "name": "RELAI; University of Maryland, College Park", "url": "https://ai.engineer/orgs/relai-university-of-maryland-college-park", "summary": "At AI Engineer World's Fair 2026, Soheil Feizi presented verifiable continual learning for AI agents. His talk examined how production failures and feedback can guide improvements across model, harness, and memory layers, comparing fine-tuning, reinforcement learning, LoRA, and prompt optimization.", "speakerSlugs": [ "soheil-feizi" ], "talkSlugs": [ "continual-learning-for-ai-agents-from-failures-to-durable-improvements-soheil-feizi-relai" ] }, { "slug": "replicate", "name": "Replicate", "url": "https://ai.engineer/orgs/replicate", "summary": "At AI Engineer World's Fair 2025, Replicate-affiliated speaker Zeke Sikelianos examined Andrej Karpathy's experience building MenuGen. His talk traces how outdated documentation, API changes, rate limiting, deployment friction, and account onboarding complicate AI-assisted development.", "speakerSlugs": [ "zeke-sikelianos" ], "talkSlugs": [ "design-like-karpathy-is-watching" ] }, { "slug": "resolve-ai", "name": "Resolve AI", "url": "https://ai.engineer/orgs/resolve-ai", "summary": "At AI Engineer World's Fair 2026, Justin Smith discussed always-on agents for production operations. His talk argues that faster AI-assisted coding leaves operational bottlenecks intact, and describes background agents intended to monitor deployments, investigate anomalies, and reduce recurring on-call work.", "speakerSlugs": [ "justin-smith" ], "talkSlugs": [ "always-on-agents-run-production-without-the-on-call-tax" ] }, { "slug": "resonate-hq-inc", "name": "Resonate HQ, Inc", "url": "https://ai.engineer/orgs/resonate-hq-inc", "summary": "At AI Engineer World's Fair 2026, Dominik Tornow presented coding agents as a way to generate bespoke software from abstract specifications and existing infrastructure. Using Resonate's durable-execution protocol, he argued that specifications could become products, replacing general-purpose platforms with implementations tailored to their deployment targets.", "speakerSlugs": [ "dominik", "dominik-tornow" ], "talkSlugs": [ "the-prompt-is-the-platform" ] }, { "slug": "results-generation", "name": "Results Generation", "url": "https://ai.engineer/orgs/results-generation", "summary": "At AI Engineer World's Fair 2026, Results Generation's affiliated speaker Jacob E. Thomas examined how historical AI personas can reproduce popular cultural portrayals instead of historically grounded reasoning. His talk uses Alexander Hamilton, Abraham Lincoln and the InCharacter benchmark to question what convincing personality imitation actually measures.", "speakerSlugs": [ "jacob-e-thomas" ], "talkSlugs": [ "the-miranda-hypothesis-how-hamilton-the-musical-poisoned-your-persona-evals" ] }, { "slug": "retool", "name": "Retool", "url": "https://ai.engineer/orgs/retool", "summary": "At AI Engineer World's Fair 2025, Retool-affiliated speaker Donald Hruska argued that practical returns from enterprise AI depend on agents with production integrations, security, and guardrails. His talk connected progress in coding agents to broader business automation.", "speakerSlugs": [ "donald-hruska" ], "talkSlugs": [ "how-agents-will-unlock-the-500b-promise-of-ai" ] }, { "slug": "risa-labs", "name": "RISA Labs", "url": "https://ai.engineer/orgs/risa-labs", "summary": "At AI Engineer World's Fair 2026, Anant Shankhdhar presented RISA Labs' automation of oncology prior authorization through collaborating agents. The talk covers insurance checks through APIs or robotic process automation, payer-data normalization, and decision corroboration using historical records and a SQL-backed payer-rule knowledge base.", "speakerSlugs": [ "anant-shankhdhar" ], "talkSlugs": [ "can-oncology-workflows-run-without-human-touch-anant-shankhdhar-risa-labs" ] }, { "slug": "rtrvr-ai", "name": "rtrvr.ai", "url": "https://ai.engineer/orgs/rtrvr-ai", "summary": "At AI Engineer Summit 2025, Arjun Chintapalli and Bhavani Kalisetty demonstrated rtrvr.ai, a Chrome-extension-based agent for natural-language browser tasks. Their examples covered product comparisons, document summarization, competitive research, and collaborative dataset creation, with structured data extraction into Google Sheets and work across multiple tabs.", "speakerSlugs": [ "arjun-chintapalli", "bhavani-kalisetty" ], "talkSlugs": [ "beyond-apis-how-ai-web-agents-are-automating-the-long-tail-of-knowledge-work" ] }, { "slug": "sakana-ai", "name": "Sakana AI", "url": "https://ai.engineer/orgs/sakana-ai", "summary": "At AI Engineer World's Fair 2026, Stefania Druga presented an on-device memory harness for research agents whose context degrades and decisions get forgotten. The talk combines persistent core traces, archival memory, vector retrieval, and decision ledgers to support longer research tasks.", "speakerSlugs": [ "stefania-druga" ], "talkSlugs": [ "memory-harnesses-for-long-running-research-agents" ] }, { "slug": "salesforce", "name": "Salesforce", "url": "https://ai.engineer/orgs/salesforce", "summary": "At AI Engineer World's Fair 2024, Salesforce-affiliated speaker Shelby Heinecke discussed efficiency as a way to move AI demonstrations into enterprise production and onto resource-constrained devices. Her talk covers efficient architectures, smaller language models, and methods for assessing on-device performance.", "speakerSlugs": [ "shelby-heinecke" ], "talkSlugs": [ "a-practical-guide-to-efficient-ai" ] }, { "slug": "scalekit", "name": "Scalekit", "url": "https://ai.engineer/orgs/scalekit", "summary": "At AI Engineer World's Fair 2026, Ravi Madabhushi examined how agent traffic exposed assumptions in APIs designed for people. His Scalekit-affiliated talk traced recurring latency spikes and database-write pressure to a last-seen timestamp, then considered identity and authorization requirements for non-deterministic agents.", "speakerSlugs": [ "ravi-madabhushi" ], "talkSlugs": [ "you-didn-t-ship-a-bug-you-just-wrote-it-for-a-human" ] }, { "slug": "scorecard", "name": "Scorecard", "url": "https://ai.engineer/orgs/scorecard", "summary": "At AI Engineer World's Fair 2025, Scorecard's Darius Emrani examined how benchmark incentives can distort AI model comparisons. His talk argued for evaluations grounded in the deployment domain and task, rather than treating benchmark performance as sufficient evidence of practical usefulness.", "speakerSlugs": [ "darius-emrani" ], "talkSlugs": [ "the-benchmarks-game-why-it-s-rigged-and-how-you-can-really-win" ] }, { "slug": "sematic", "name": "Sematic", "url": "https://ai.engineer/orgs/sematic", "summary": "At AI Engineer Summit 2023, Sematic's Emmanuel Turlay examines how to evaluate language models for a particular application. He explains the limits of generic benchmarks and proposes using an LLM to grade outputs against criteria drawn from the intended use case.", "speakerSlugs": [ "emmanuel-turlay" ], "talkSlugs": [ "how-to-evaluate-a-model-for-your-use-case" ] }, { "slug": "sf-compute", "name": "SF Compute", "url": "https://ai.engineer/orgs/sf-compute", "summary": "At AI Engineer World's Fair 2025, John Pham used SF Compute’s GPU-cloud onboarding to show how interface performance shapes perceived quality. His talk argues that faster software development with AI leaves the human-centered fundamentals of good design intact.", "speakerSlugs": [ "john-pham" ], "talkSlugs": [ "good-design-hasn-t-changed-with-ai" ] }, { "slug": "signalfire", "name": "SignalFire", "url": "https://ai.engineer/orgs/signalfire", "summary": "At AI Engineer Summit 2025, SignalFire’s Heath Black discussed using Beacon AI talent intelligence to inform AI-team hiring. His talk emphasized demonstrated engineering experience, open-source work, and applied ML and MLOps skills over reliance on elite-school or PhD credentials.", "speakerSlugs": [ "heath-black" ], "talkSlugs": [ "insights-on-building-ai-teams" ] }, { "slug": "sky-valley-ambient-computing", "name": "Sky Valley Ambient Computing", "url": "https://ai.engineer/orgs/sky-valley-ambient-computing", "summary": "At AI Engineer World's Fair 2026, Iris ten Teije presented Differ, Sky Valley Ambient Computing's adaptive-software product. Her talk challenges the assumption that every user receives one frozen artifact, describing a shared software stem with bounded, isolated, reversible changes for individual users.", "speakerSlugs": [ "iris-ten-teije" ], "talkSlugs": [ "the-pipeline-is-dead" ] }, { "slug": "smithery", "name": "Smithery", "url": "https://ai.engineer/orgs/smithery", "summary": "At AI Engineer World's Fair 2025, Henry Mao argues that stronger models and the Model Context Protocol alone cannot make agents practically useful. He presents Smithery as an orchestration gateway and demonstrates an agent connecting GitHub work to Linear.", "speakerSlugs": [ "henry-mao" ], "talkSlugs": [ "are-mcps-overhyped-a-rant-about-mcps" ] }, { "slug": "snapchat", "name": "Snapchat", "url": "https://ai.engineer/orgs/snapchat", "summary": "Jeffrey Lee-Chan’s OpenClaw workshop at AI Engineer World's Fair 2026 examines development with parallel coding agents. The session covers persistent Slack-derived memory, isolated worktrees, and tmux-managed terminals, with audience questions extending to multi-model workflows and Docker-based sandboxing.", "speakerSlugs": [ "jeffrey-lee-chan" ], "talkSlugs": [ "develop-at-idea-velocity" ] }, { "slug": "snaplogic-university-of-san-francisco", "name": "SnapLogic; University of San Francisco", "url": "https://ai.engineer/orgs/snaplogic-university-of-san-francisco", "summary": "At AI Engineer World's Fair 2025, Greg Benson presented Agent Continuations: a mechanism for suspending and resuming LLM workflows without keeping agent loops running. His talk connects this approach to human approval and failure recovery, and introduces a public prototype using the OpenAI Python API.", "speakerSlugs": [ "greg-benson" ], "talkSlugs": [ "breaking-the-chain-agent-continuations-for-resumable-ai-workflows" ] }, { "slug": "snorkel", "name": "Snorkel", "url": "https://ai.engineer/orgs/snorkel", "summary": "At AI Engineer World's Fair 2024, Snorkel's Humza Iqbal joined Microsoft's Lachlan Ainley to discuss curated, programmatic data development for enterprise AI beyond off-the-shelf models and basic RAG. Their discussion connected expert feedback and domain-specific evaluation with practical deployment on Azure AI infrastructure.", "speakerSlugs": [ "humza-iqbal" ], "talkSlugs": [ "insights-from-snorkel-ai-running-azure-ai-infrastructure" ] }, { "slug": "snowglobe", "name": "Snowglobe", "url": "https://ai.engineer/orgs/snowglobe", "summary": "At AI Engineer, Snowglobe’s Shreya Rajpal joins Nubank’s Aman Gupta to discuss simulation-based evaluation for customer-support agents. Their joint session examines bottlenecks in hand-curated evaluations and production A/B tests, alongside a reported 20× acceleration in deployment.", "speakerSlugs": [ "shreya-rajpal" ], "talkSlugs": [ "simulation-maxxing-how-nubank-ships-agents-20-faster-with-simulations" ] }, { "slug": "software-3-0-llc", "name": "Software 3.0, LLC", "url": "https://ai.engineer/orgs/software-3-0-llc", "summary": "Benjamin Dunphy introduced Network, the conference mobile application, during the opening of AI Engineer Summit 2023. This recorded appearance connects Software 3.0, LLC to a practical use of LLMs for matching conference attendees through the app’s Generative Matching Algorithm.", "speakerSlugs": [ "benjamin-dunphy" ], "talkSlugs": [ "announcing-the-ai-engineer-network" ] }, { "slug": "sondermind", "name": "SonderMind", "url": "https://ai.engineer/orgs/sondermind", "summary": "At AI Engineer World's Fair 2026, Akele Reed and Dave Revere explain the engineering behind Sonder, SonderMind's clinically grounded AI mental-health coach. Their presentation describes a system designed to complement therapy and direct users to human care when appropriate.", "speakerSlugs": [ "akele-reed", "dave-revere", "doug-keller" ], "talkSlugs": [ "evals-driven-development-engineering-a-mental-health-ai-coach-ethically-safely" ] }, { "slug": "specstory", "name": "SpecStory", "url": "https://ai.engineer/orgs/specstory", "summary": "At AI Engineer Summit 2025, SpecStory's Greg Ceccarelli joined Hamel Husain of Parlance Labs for a satirical discussion of enterprise AI strategy failures. The session examines how executive isolation, hype-driven priorities, and unrealistic spending can undermine work on concrete business problems.", "speakerSlugs": [ "greg-ceccarelli" ], "talkSlugs": [ "how-to-build-an-ai-strategy-that-fails" ] }, { "slug": "spotify", "name": "Spotify", "url": "https://ai.engineer/orgs/spotify", "summary": "At AI Engineer World's Fair 2025, Shivam Verma presented Spotify's approach to steerable music and podcast recommendations using listener representations and language models. The talk connects that architecture to Discover Weekly, DJ, and Prompted Playlist.", "speakerSlugs": [ "shivam-verma" ], "talkSlugs": [ "personalization-in-the-era-of-llms" ] }, { "slug": "sprout-social", "name": "Sprout Social", "url": "https://ai.engineer/orgs/sprout-social", "summary": "At AI Engineer World's Fair 2025, Rossella Blatt Vital and Deepsha Menghani discuss moving an established SaaS business beyond isolated AI features while continuing roadmap delivery. Their talk outlines a transformation framework spanning organizational processes, company-wide AI fluency, and T-shaped technical talent.", "speakerSlugs": [ "rossella-blatt-vital", "deepsha-menghani" ], "talkSlugs": [ "from-hype-to-habit-how-we-re-building-an-ai-first-saas-company-while-still-shipping-the-roadmap" ] }, { "slug": "standardagents", "name": "StandardAgents", "url": "https://ai.engineer/orgs/standardagents", "summary": "At AI Engineer World's Fair 2026, Justin Schroeder presented a case for composing narrowly scoped, domain-specific agents instead of loading a general-purpose agent with more tools, MCP servers, and skills. His talk also considered specialized legal and GDPR-compliance agents.", "speakerSlugs": [ "justin-schroeder" ], "talkSlugs": [ "the-future-is-domain-specific-agents" ] }, { "slug": "starlightsearch-inc", "name": "StarlightSearch Inc", "url": "https://ai.engineer/orgs/starlightsearch-inc", "summary": "At AI Engineer World's Fair 2026, Sonam Pankaj presented Reflect, a runtime memory layer designed to bring evaluation outcomes into future retrieval. Her talk examines repeated failures in ReAct-style agents and demonstrates failure feedback through a SQL product-search agent.", "speakerSlugs": [ "sonam-pankaj" ], "talkSlugs": [ "user-signal-die-at-the-retrieval-boundary" ] }, { "slug": "stride", "name": "Stride", "url": "https://ai.engineer/orgs/stride", "summary": "At AI Engineer World's Fair 2025, Stride's Dan Mason presented a telemedicine support system developed with Avila Science for multi-day, at-home treatment. The workshop covered patient support that preserves clinician relationships and human escalation, alongside LangGraph orchestration and MCP-connected tools.", "speakerSlugs": [ "dan-mason" ], "talkSlugs": [ "case-study-deep-dive-telemedicine-support-agents-with-langgraph-mcp" ] }, { "slug": "substrate", "name": "Substrate", "url": "https://ai.engineer/orgs/substrate", "summary": "At AI Engineer World's Fair 2024, Rob Cheung introduced Substrate through multimodal computation graphs that connect image generation, speech transcription, text, embeddings, and code execution. His presentation paired a developer SDK with an inference engine designed to execute those graphs efficiently.", "speakerSlugs": [ "rob-cheung" ], "talkSlugs": [ "substrate-launch-the-api-for-modular-ai" ] }, { "slug": "super-protocol", "name": "Super Protocol", "url": "https://ai.engineer/orgs/super-protocol", "summary": "At AI Engineer World's Fair 2025, Mike Bursell presented Super Protocol's confidential AI cloud, explaining hardware-backed trusted execution environments and cryptographic attestation for protecting data and models during training, fine-tuning, and inference. The presentation also covered healthcare data access, model provenance, and decentralized AI marketplaces.", "speakerSlugs": [ "mike-bursell" ], "talkSlugs": [ "gpu-less-trust-less-limit-less-reimagining-the-confidential-ai-cloud" ] }, { "slug": "superagentic-ai", "name": "Superagentic AI", "url": "https://ai.engineer/orgs/superagentic-ai", "summary": "At AI Engineer World's Fair 2026, Shashi Jagtap presented Recursive Language Models as an alternative for coding agents struggling with large repositories. His demonstration kept repository context in a programmable REPL, using generated code to inspect it and recursive model calls to answer focused questions.", "speakerSlugs": [ "shashi-jagtap" ], "talkSlugs": [ "rlm-recursive-language-models-for-large-codebases" ] }, { "slug": "superconductor", "name": "Superconductor", "url": "https://ai.engineer/orgs/superconductor", "summary": "At AI Engineer World's Fair 2026, Arjun Singh presented engineering collaboration through persistent coding-agent sessions spanning Slack, GitHub, desktop, and mobile. His talk connects these shared workflows with automated conversion of meetings and customer feedback into pull requests.", "speakerSlugs": [ "arjun-singh" ], "talkSlugs": [ "multiplayer-agentic-engineering-enabling-your-whole-team-and-your-best-agents-to-work-together" ] }, { "slug": "superdial", "name": "SuperDial", "url": "https://ai.engineer/orgs/superdial", "summary": "At AI Engineer Summit 2025, SuperDial's affiliated speaker Nik Caryotakis presented production voice agents for healthcare insurance calls. His talk emphasizes conversation design, phone-tree navigation, EHR integrations, and operational reliability over voice realism, illustrated through an insurance-call workflow.", "speakerSlugs": [ "nik-caryotakis" ], "talkSlugs": [ "voice-ai-your-bot-isn-t-special" ] }, { "slug": "superintelligent", "name": "Superintelligent", "url": "https://ai.engineer/orgs/superintelligent", "summary": "At AI Engineer Code 2025, Nathaniel Whittemore (NLW) examined enterprise AI value, contrasting rising agent deployment with difficulty scaling beyond pilots. His talk covered executive spending and ROI expectations, alongside reported time savings and differences between adoption leaders and laggards.", "speakerSlugs": [ "nathaniel-whittemore-nlw" ], "talkSlugs": [ "ai-consulting-in-practice-nathaniel-whittemore-nlw-superintelligent" ] }, { "slug": "superlinked", "name": "Superlinked", "url": "https://ai.engineer/orgs/superlinked", "summary": "At AI Engineer World's Fair 2025, Filip Makraduli presented infrastructure for small models powering embeddings, retrieval, document processing, and agentic workflows. His talk introduced Superlinked Inference Engine, connecting model support with production routing, autoscaling, queuing, and GPU provisioning.", "speakerSlugs": [ "filip-makraduli" ], "talkSlugs": [ "the-small-model-infrastructure-nobody-built-so-we-did-filip-makraduli-superlinked" ] }, { "slug": "surge-ai", "name": "Surge AI", "url": "https://ai.engineer/orgs/surge-ai", "summary": "At AI Engineer World's Fair 2026, Nick Heiner of Surge AI examines why benchmark optimization can diverge from practical AI usefulness. His talk connects leaderboard incentives and public-data contamination with evaluation failures, using examples from LMArena, SWE-bench Verified, and IFEval.", "speakerSlugs": [ "nick-heiner" ], "talkSlugs": [ "when-will-the-benchmaxxing-plague-end" ] }, { "slug": "synth", "name": "Synth", "url": "https://ai.engineer/orgs/synth", "summary": "At AI Engineer World's Fair 2025, Josh Purtell presented stateful environments for vertical agents working over long horizons. His talk describes how simplified representations of external systems such as Excel can preserve useful state, drawing on reinforcement-learning environments and agent benchmarks.", "speakerSlugs": [ "josh-purtell" ], "talkSlugs": [ "stateful-environments-for-vertical-agents-josh-purtell-synth-labs" ] }, { "slug": "tailscale", "name": "Tailscale", "url": "https://ai.engineer/orgs/tailscale", "summary": "At AI Engineer Europe 2026, Remy Guercio examined agent sandboxing by separating execution isolation from credential and access control. His talk connected Tailscale's identity layer over WireGuard with a demonstration of Aperture, an identity-aware AI gateway for developers and automated agents.", "speakerSlugs": [ "remy-guercio" ], "talkSlugs": [ "what-if-the-network-was-the-sandbox" ] }, { "slug": "take-take-take", "name": "Take Take Take", "url": "https://ai.engineer/orgs/take-take-take", "summary": "At AI Engineer Europe 2026, Anant Dole and Asbjørn Steinskog presented Take Take Take's consumer chess coach, combining Stockfish position analysis and grounded chess signals with language-model explanations. Their talk examines why general-purpose LLMs struggle to play chess reliably.", "speakerSlugs": [ "anant-dole", "asbj-rn-steinskog" ], "talkSlugs": [ "building-a-chess-coach" ] }, { "slug": "taste-labs", "name": "Taste Labs", "url": "https://ai.engineer/orgs/taste-labs", "summary": "At AI Engineer World's Fair 2026, Thais Castello Branco presented ways to reduce generic AI-generated creative work by translating subjective qualities into measurable, context-specific signals. Her talk uses Reducto's brand to illustrate how design attributes can become verifiable reinforcement-learning tasks.", "speakerSlugs": [ "thais-castello-branco" ], "talkSlugs": [ "ending-ai-slop" ] }, { "slug": "tavily", "name": "Tavily", "url": "https://ai.engineer/orgs/tavily", "summary": "Evaluating AI-search agents against changing web content is the focus of Maitar Asher’s recorded Tavily appearance at AI Engineer World's Fair 2025. In a joint session with Julia Neagu and Deanna Emery of Quotient AI, the discussion addresses unpredictable queries and production evaluation.", "speakerSlugs": [ "maitar-asher" ], "talkSlugs": [ "evaluating-ai-search-a-practical-framework-for-augmented-ai-systems" ] }, { "slug": "tavon-ai", "name": "TAVON.ai", "url": "https://ai.engineer/orgs/tavon-ai", "summary": "At AI Engineer Europe 2026, Matthias Luebken explains how to embed Pi, the minimal coding-agent toolkit used by OpenClaw, into business products. His demonstrations connect CRM lead qualification and inbox-driven sales workflows with TypeScript extensions and interactive controls.", "speakerSlugs": [ "matthias-luebken" ], "talkSlugs": [ "a-piece-of-pi-embedding-the-openclaw-coding-agent-in-your-product" ] }, { "slug": "tavus", "name": "Tavus", "url": "https://ai.engineer/orgs/tavus", "summary": "Realtime conversational video agents are the focus of Brian Johnson of Tavus and Chad Bailey of Daily's joint session at AI Engineer World's Fair 2025. Their discussion connects voice-model pipelines and Pipecat orchestration with Tavus digital-human video.", "speakerSlugs": [ "brian-johnson" ], "talkSlugs": [ "realtime-conversational-video-with-pipecat-and-tavus-chad-bailey-and-brian-johnson-daily-tavus" ] }, { "slug": "teammates", "name": "Teammates", "url": "https://ai.engineer/orgs/teammates", "summary": "At AI Engineer World's Fair 2025, Ben Stein of Teammates discussed building products around unpredictable LLM capabilities. His talk uses a virtual colleague working across email, Slack, and document comments to explain why autonomous AI coworkers need broad affordances instead of rigid feature definitions.", "speakerSlugs": [ "ben-stein" ], "talkSlugs": [ "shipping-products-when-you-don-t-know-what-they-can-do" ] }, { "slug": "telemetrak", "name": "Telemetrak", "url": "https://ai.engineer/orgs/telemetrak", "summary": "At AI Engineer World's Fair 2025, Telemetrak-affiliated speakers Sahil Yadav and Hariharan Ganesan examined trustworthy AI inference in telecom, industrial IoT, and safety-critical environments. Their talk connected plain-language explainability, adaptive guardrails with human escalation, and digitally signed traceability to operational oversight.", "speakerSlugs": [ "sahil-yadav", "hariharan-ganesan" ], "talkSlugs": [ "critical-ai-inference-your-cio-can-trust" ] }, { "slug": "tenex", "name": "Tenex", "url": "https://ai.engineer/orgs/tenex", "summary": "At AI Engineer Code 2025, Arman Hezarkhani presented a compensation model linking client charges to delivered story points and engineer pay to completed story points. His talk argues that traditional compensation discourages AI adoption and examines incentives for engineering delivery.", "speakerSlugs": [ "arman-hezarkhani" ], "talkSlugs": [ "paying-engineers-like-salespeople" ] }, { "slug": "tesco", "name": "Tesco", "url": "https://ai.engineer/orgs/tesco", "summary": "At AI Engineer World's Fair 2026, Tesco-affiliated speaker Rajkumar Sakthivel presented Code Context Engine, a local retrieval layer for AI coding tools. The talk describes how indexed snippets, cross-file relationships, and shared project memory reduce unnecessary code context sent to models.", "speakerSlugs": [ "rajkumar-sakthivel" ], "talkSlugs": [ "we-cut-94-of-our-ai-coding-tokens-with-a-local-code-index-here-s-the-architecture" ] }, { "slug": "tesla", "name": "Tesla", "url": "https://ai.engineer/orgs/tesla", "summary": "At AI Engineer World's Fair 2026, Tesla-affiliated speaker Ishita Daga examined enterprise data agents' difficulty interpreting ambiguous or stale business context. Her talk argues that clearer organizational knowledge, rather than larger models, is central to making these agents reliable.", "speakerSlugs": [ "ishita-daga" ], "talkSlugs": [ "enterprise-agents-have-a-structure-problem-ishita-daga-tesla" ] }, { "slug": "tesla-optimus", "name": "Tesla Optimus", "url": "https://ai.engineer/orgs/tesla-optimus", "summary": "At AI Engineer World's Fair 2025, Rishabh Garg examined robotics failures that appear to stem from control policies but originate in software, communication, and timing. His talk used a toy robot architecture to explain how infrastructure problems disrupt reliable control.", "speakerSlugs": [ "rishabh-garg" ], "talkSlugs": [ "challenges-in-high-performance-robotics-systems" ] }, { "slug": "tessl", "name": "Tessl", "url": "https://ai.engineer/orgs/tessl", "summary": "At AI Engineer Europe 2026, Patrick Debois of Tessl presented context for AI coding agents as an engineering concern. His talk covers the instructions, skills, documentation, and organizational knowledge that guide agents, organized around a Context Development Lifecycle spanning generation, evaluation, distribution, and observation.", "speakerSlugs": [ "patrick-debois" ], "talkSlugs": [ "context-is-the-new-code" ] }, { "slug": "the-new-york-times", "name": "The New York Times", "url": "https://ai.engineer/orgs/the-new-york-times", "summary": "At AI Engineer World's Fair 2026, Shafik Quoraishee and Joanne Song presented experimental on-device agents for accessible mobile games, including a Space Invaders prototype. They explored gameplay adaptations for eye gaze, shaky taps, low vision, and handwriting.", "speakerSlugs": [ "shafik-quoraishee", "joanne-song" ], "talkSlugs": [ "local-agentic-theory-for-mobile-games-shafik-quoraishee-joanne-song-the-new-york-times" ] }, { "slug": "the-new-york-times-games", "name": "The New York Times Games", "url": "https://ai.engineer/orgs/the-new-york-times-games", "summary": "At AI Engineer World's Fair 2025, Shafik Quoraishee presented independent, preliminary research into Connections as a benchmark for AI reasoning. The talk examines how decoys, semantic ambiguity, and words with multiple meanings challenge language models solving word puzzles.", "speakerSlugs": [ "shafik-quoraishee" ], "talkSlugs": [ "new-york-times-connections-a-case-study-on-nlp-in-word-games" ] }, { "slug": "the-tree-center", "name": "The Tree Center", "url": "https://ai.engineer/orgs/the-tree-center", "summary": "At AI Engineer World's Fair 2024, The Tree Center's affiliated speaker Manuel Odendahl led a hands-on workshop on LLMs for everyday software development. The session covered lightweight model setup, GitHub-based developer onboarding, zero-shot prompting, and practical Copilot use.", "speakerSlugs": [ "manuel-odendahl" ], "talkSlugs": [ "llms-for-the-working-programmer-become-a-10x-programming-centaur-today" ] }, { "slug": "thebrain-pro", "name": "TheBrain.pro", "url": "https://ai.engineer/orgs/thebrain-pro", "summary": "At AI Engineer World's Fair 2025, Łukasz Gandecki demonstrated BookGenius, a reading experience that adds character context, avatars, music, and embedding-based scene search. The presentation used The Snow Queen and 1984 to show how AI can extend familiar books.", "speakerSlugs": [ "ukasz-gandecki" ], "talkSlugs": [ "books-reimagined-ai-to-create-new-experiences-for-things-you-know-ukasz-gandecki-thebrain-pro" ] }, { "slug": "theory-ventures", "name": "Theory Ventures", "url": "https://ai.engineer/orgs/theory-ventures", "summary": "At AI Engineer Summit 2025, Andy Triedman of Theory Ventures presented a framework for choosing enterprise jobs for LLM automation. The talk connects transformation, synthesis, and reasoning capabilities to workflow design, using security alert investigation and personalization as examples.", "speakerSlugs": [ "andy-triedman" ], "talkSlugs": [ "where-ai-is-superhuman-the-right-jobs-to-automate-with-llms" ] }, { "slug": "theta-software", "name": "Theta Software", "url": "https://ai.engineer/orgs/theta-software", "summary": "At AI Engineer World's Fair 2026, Theta Software's Rayan Garg presented on defining and evaluating long-horizon AI-agent work. The talk contrasts human-time success thresholds with token- and trajectory-based measures, explaining why human effort can misrepresent model difficulty.", "speakerSlugs": [ "rayan-garg" ], "talkSlugs": [ "rethinking-environments-for-long-horizon-work" ] }, { "slug": "thinking-machines", "name": "Thinking Machines", "url": "https://ai.engineer/orgs/thinking-machines", "summary": "Frontier-model training, coding agents, and on-device deployment feature in Frontier Feud, where Steven participated with a Thinking Machines affiliation at AI Engineer Summit 2025. Hosted by Barr Yaron, the shared session uses a Family Feud-style contest to surface participants’ predictions.", "speakerSlugs": [ "steven" ], "talkSlugs": [ "frontier-feud" ] }, { "slug": "thryv-com", "name": "thryv.com", "url": "https://ai.engineer/orgs/thryv-com", "summary": "At AI Engineer World's Fair 2025, thryv.com's affiliated speaker Brook Riggio demonstrated a serverless AI application combining Next.js, the OpenAI Agents SDK, Inngest, and Python functions on Vercel. His newsletter-generation demo turned user-supplied topics into an orchestrated agent workflow.", "speakerSlugs": [ "brook-riggio" ], "talkSlugs": [ "the-demo-i-wish-i-d-had-openai-s-agents-sdk-serverless" ] }, { "slug": "tinder", "name": "Tinder", "url": "https://ai.engineer/orgs/tinder", "summary": "At AI Engineer World's Fair 2024, Vibhor Kumar described how generative AI both amplifies spam and scams and enables stronger harm detection on Tinder. His talk covers classification datasets built through human/LLM workflows and adaptation to adversarial fraud.", "speakerSlugs": [ "vibhor-kumar" ], "talkSlugs": [ "ai-frontiers-in-trust-and-safety-combatting-multifaceted-harm-on-tinder-at-scale" ] }, { "slug": "tng-technology-consulting", "name": "TNG Technology Consulting", "url": "https://ai.engineer/orgs/tng-technology-consulting", "summary": "At AI Engineer Europe 2026, TNG Technology Consulting’s Stephan Steinfurt demonstrates an autonomous workflow that turns nightly Lichess game downloads into narrated chess puzzle videos. The talk follows game analysis through automatic YouTube publication and includes audience questions about video quality and cost.", "speakerSlugs": [ "stephan-steinfurt" ], "talkSlugs": [ "running-a-chess-youtube-channel-entirely-by-ai-stephan-steinfurt-tng-technology-consulting" ] }, { "slug": "tola-capital", "name": "Tola Capital", "url": "https://ai.engineer/orgs/tola-capital", "summary": "At AI Engineer World's Fair 2024, Tola Capital's Sheila Gulati joined Klarity's Nischal Nadhamuni to discuss evaluations for increasingly agentic AI systems. Their discussion connects evaluation to operational objectives, human values, and real-world performance.", "speakerSlugs": [ "sheila-gulati" ], "talkSlugs": [ "e-values-evaluating-the-values-of-ai" ] }, { "slug": "trainline", "name": "Trainline", "url": "https://ai.engineer/orgs/trainline", "summary": "At AI Engineer Europe 2026, Oussama Hafferssas and Mayank Soni joined Braintrust’s Giran Moodley for a hands-on workshop on reliable AI applications. The discussion covered Trainline’s agentic systems, splitting LLM workflows into tool-enabled specialist stages, and diagnosing failures through application tracing.", "speakerSlugs": [ "oussama-hafferssas", "mayank-soni" ], "talkSlugs": [ "shipping-complex-ai-applications-braintrust-trainline" ] }, { "slug": "trajectory", "name": "Trajectory", "url": "https://ai.engineer/orgs/trajectory", "summary": "At AI Engineer World's Fair 2026, Trajectory's Ronak Malde argued for turning real-world model interactions into continual-learning training signals. His talk contrasted this approach with reliance on increasingly costly, disconnected benchmarks and addressed distribution shifts, reward hacking, and continual-training latency.", "speakerSlugs": [ "ronak-malde" ], "talkSlugs": [ "scaling-up-continual-learning" ] }, { "slug": "traversal", "name": "Traversal", "url": "https://ai.engineer/orgs/traversal", "summary": "At AI Engineer World's Fair 2025, Anish Agarwal and Matthew Schoenbauer argued that AI-generated software increases production complexity while reducing engineers' operational context. Their Traversal presentation described autonomous troubleshooting and used a DigitalOcean case study to illustrate incident investigation and reduced resolution time.", "speakerSlugs": [ "anish-agarwal", "matthew-schoenbauer" ], "talkSlugs": [ "production-software-keeps-breaking-and-it-will-only-get-worse-here-s-how-traversal-is-fixing-it" ] }, { "slug": "trigger-dev", "name": "Trigger.dev", "url": "https://ai.engineer/orgs/trigger-dev", "summary": "At AI Engineer World's Fair 2025, Eric Allam compared replay-based durable execution with snapshot infrastructure for long-running AI agents. His talk examined why recovery requires preserving both LLM conversation context and execution environments containing files, memory, subprocesses, and tools.", "speakerSlugs": [ "eric-allam" ], "talkSlugs": [ "two-roads-to-durable-agents-replay-vs-snapshot-eric-allam-co-founder-trigger-dev" ] }, { "slug": "trunk-tools", "name": "Trunk Tools", "url": "https://ai.engineer/orgs/trunk-tools", "summary": "At AI Engineer World's Fair 2024, Dr. Sarah Buchner presented Trunk Tools' construction-specific retrieval-augmented generation for large project document collections. Her demonstration focused on source-grounded question answering, using a skyscraper project with 3.6 million pages to illustrate the scale.", "speakerSlugs": [ "dr-sarah-buchner" ], "talkSlugs": [ "trunk-tools-launch-disrupting-the-15-trillion-construction-industry-with-autonomous-agents" ] }, { "slug": "turbopuffer", "name": "turbopuffer", "url": "https://ai.engineer/orgs/turbopuffer", "summary": "At AI Engineer Europe 2026, Kuba Rogut benchmarks semantic code retrieval in Claude Code against conventional grep-based exploration. The talk compares baseline Claude Code, 50-line windowed reads, and windowed reads augmented with semantic search, giving readers a concrete experimental setup for examining coding-agent retrieval.", "speakerSlugs": [ "kuba-rogut" ], "talkSlugs": [ "benchmarking-semantic-code-retrieval-on-claude-code" ] }, { "slug": "tusk", "name": "Tusk", "url": "https://ai.engineer/orgs/tusk", "summary": "At AI Engineer World's Fair 2025, Tusk's Jun Yu Tan discussed designing AI interfaces to strengthen human judgment in complex work. Drawing on email composition and code review, he argued that effective augmentation should develop user capabilities and measure growth alongside product engagement.", "speakerSlugs": [ "jun-yu-tan" ], "talkSlugs": [ "designing-ai-to-scale-human-thought" ] }, { "slug": "twelvelabs", "name": "TwelveLabs", "url": "https://ai.engineer/orgs/twelvelabs", "summary": "At AI Engineer World's Fair 2026, James Le presented durable, configurable memory for video AI as an alternative to answering each query from scratch. His demonstrations included tracking Lionel Messi across footage, traffic-safety detection, and advertising use cases.", "speakerSlugs": [ "james-le" ], "talkSlugs": [ "video-has-no-memory-here-s-how-we-built-one" ] }, { "slug": "typesafe-ai", "name": "TypeSafe AI", "url": "https://ai.engineer/orgs/typesafe-ai", "summary": "At AI Engineer World's Fair 2026, Diogo Almeida examined how RLHF's emphasis on pleasing humans and interactive assistance can constrain reliable autonomous decisions. His talk calls for software designed to perform consequential work without continuous human oversight.", "speakerSlugs": [ "diogo-almeida" ], "talkSlugs": [ "what-s-next-after-rlhf" ] }, { "slug": "ufonia", "name": "Ufonia", "url": "https://ai.engineer/orgs/ufonia", "summary": "At AI Engineer World's Fair 2026, Jared Joselowitz presents safety evaluation for Dora, a regulated clinical voice agent. His talk explains why conventional A/B testing, incremental rollout, and rollback cannot adequately protect patients, then introduces MATRIX, a hazard-driven simulation framework for evaluation before deployment.", "speakerSlugs": [ "jared-joselowitz" ], "talkSlugs": [ "shipping-ai-to-a-million-patients-without-an-a-b-test" ] }, { "slug": "untapped-capital", "name": "Untapped Capital", "url": "https://ai.engineer/orgs/untapped-capital", "summary": "At AI Engineer World's Fair 2026, Yohei Nakajima of Untapped Capital presents ActiveGraph, an experimental runtime developed from his BabyAGI and graph-agent work. The talk centers agent activity on an immutable event log and shared graph state, enabling auditability, replay, rollback, and forking.", "speakerSlugs": [ "yohei-nakajima" ], "talkSlugs": [ "active-graph-agent-runtime-babyagi-4" ] }, { "slug": "urun", "name": "uRun", "url": "https://ai.engineer/orgs/urun", "summary": "At AI Engineer World's Fair 2026, uRun's Keegan McCallum discussed generative video's progress toward efficient, continuous, real-time output alongside visual quality. Using Helios and its Wan 2.1 14B lineage, the talk covered long-horizon generation, declining usage costs, interactive avatars, and accessibility.", "speakerSlugs": [ "keegan-mccallum" ], "talkSlugs": [ "generative-video-at-the-speed-of-light" ] }, { "slug": "varick-agents", "name": "Varick Agents", "url": "https://ai.engineer/orgs/varick-agents", "summary": "At AI Engineer World's Fair 2026, Vasuman Moza described understanding and redesigning customer-specific workflows as enterprise AI's principal bottleneck. His talk connected forward-deployed engineering with mapping finance and operational processes, placing workflow discovery at the center of agent development.", "speakerSlugs": [ "vasuman-moza" ], "talkSlugs": [ "ai-tools-for-forward-deployed-engineering" ] }, { "slug": "vellum", "name": "Vellum", "url": "https://ai.engineer/orgs/vellum", "summary": "At AI Engineer Summit 2025, Vellum's Anita Kirkovska presented test-driven development as a way to improve production reliability for nondeterministic AI agents and workflows. The talk also surveyed reasoning and retrieval approaches and introduced Vellum's code-oriented Workflows SDK.", "speakerSlugs": [ "anita-kirkovska" ], "talkSlugs": [ "ai-agents-meet-test-driven-development" ] }, { "slug": "vibe-kanban", "name": "Vibe Kanban", "url": "https://ai.engineer/orgs/vibe-kanban", "summary": "At AI Engineer Europe 2026, Louis Knight-Webb argued that coding agents are moving software engineering toward planning and reviewing generated work. His talk contrasts workflows centered on upfront plans with those centered on review, and includes reflections on startup strategy and enterprise sales.", "speakerSlugs": [ "louis-knight-webb" ], "talkSlugs": [ "software-engineering-is-becoming-plan-and-review" ] }, { "slug": "viktor", "name": "Viktor", "url": "https://ai.engineer/orgs/viktor", "summary": "At AI Engineer Europe 2026, Fryderyk Wiatrowski describes Viktor’s evolution from browser and email agents into an AI coworker inside Slack. The talk examines the engineering challenges of giving an organization a shared agent with access to company tools.", "speakerSlugs": [ "fryderyk-wiatrowski" ], "talkSlugs": [ "viktor-ai-coworker-that-lives-in-slack" ] }, { "slug": "visuallabs", "name": "VisualLabs", "url": "https://ai.engineer/orgs/visuallabs", "summary": "At AI Engineer World's Fair 2026, VisualLabs speaker Balázs Horváth examined how to choose valuable problems for AI agents. His talk argues that stakeholder engagement and requirements gathering become more important as AI makes code easier to produce.", "speakerSlugs": [ "balazs-horvath" ], "talkSlugs": [ "you-can-t-prompt-the-room-the-last-skill-ai-won-t-replace" ] }, { "slug": "wandero-ai", "name": "Wandero AI", "url": "https://ai.engineer/orgs/wandero-ai", "summary": "At AI Engineer World's Fair 2026, Wandero AI’s affiliated speaker Raphael Kalandadze discusses operating AI agents after launch. His talk addresses failures in real customer conversations that conventional tests and logs can miss, alongside dashboards for tracking costs, success rates, and emerging patterns.", "speakerSlugs": [ "raphael-kalandadze" ], "talkSlugs": [ "the-missing-layer-after-launch" ] }, { "slug": "wasp", "name": "Wasp", "url": "https://ai.engineer/orgs/wasp", "summary": "At AI Engineer Summit 2023, Wasp’s Matija Sosic demonstrates Mage, a GPT-powered generator that turns a short description into a downloadable full-stack application. The demonstration follows a task-management app through database inspection and Fly.io deployment, using React, Node.js, Prisma, and Wasp.", "speakerSlugs": [ "matija-sosic" ], "talkSlugs": [ "gpt-web-app-generator-10-000-apps-created-in-a-month-matija-sosic" ] }, { "slug": "watershed-technology-inc", "name": "Watershed Technology Inc.", "url": "https://ai.engineer/orgs/watershed-technology-inc", "summary": "At AI Engineer World's Fair 2026, Andrew Dumit presented sustainability calculations as a test of an agent’s reasoning process, not just its final answer. His talk describes coding agents working with complex emissions graphs and producing artifacts that nonprogrammer experts can inspect.", "speakerSlugs": [ "andrew-dumit" ], "talkSlugs": [ "respect-the-process" ] }, { "slug": "waymo", "name": "Waymo", "url": "https://ai.engineer/orgs/waymo", "summary": "At AI Engineer World's Fair 2025, Jyh-Jing Hwang presented EMMA, which adapts Gemini for autonomous driving. The talk explains how camera inputs and navigation instructions become planned trajectories, with reasoning about unusual, safety-critical road situations.", "speakerSlugs": [ "jyh-jing-hwang" ], "talkSlugs": [ "waymo-s-emma-teaching-cars-to-think-jyh-jing-hwang-waymo" ] }, { "slug": "waypoint-ai", "name": "Waypoint AI", "url": "https://ai.engineer/orgs/waypoint-ai", "summary": "At AI Engineer Europe 2026, Waypoint AI’s affiliated speaker Šimon Podhajský demonstrated a personal AI observer that combines digital traces into weekly reflections and suggestions for reading discussions. The system reads across sources and writes analyses to an isolated review workspace.", "speakerSlugs": [ "simon-podhajsky" ], "talkSlugs": [ "cognitive-exhaust-fumes-or-read-only-ai-is-underrated-simon-podhajsky-head-of-ai-waypoint" ] }, { "slug": "weco-ai", "name": "Weco AI", "url": "https://ai.engineer/orgs/weco-ai", "summary": "At AI Engineer World's Fair 2026, Zhengyao Jiang described Weco AI's autonomous research agent Aiden participating in OpenAI's Parameter Golf challenge. His talk follows how Aiden ingested public research, ran experiments, and used pull-request quality gates in machine-learning research.", "speakerSlugs": [ "zhengyao-jiang" ], "talkSlugs": [ "how-autoresearch-is-changing-ml-research-zhengyao-jiang-weco-ai" ] }, { "slug": "weka", "name": "WEKA", "url": "https://ai.engineer/orgs/weka", "summary": "At AI Engineer Code 2025, WEKA presenters Val Bercovici and Callan Fox introduce context-platform engineering for agent swarms. Their talk covers an open-source load-generation toolkit, KV-cache optimization, and translating agent service-level agreements into infrastructure service-level objectives.", "speakerSlugs": [ "val-bercovici", "callan-fox" ], "talkSlugs": [ "context-platform-engineering-to-reduce-token-anxiety-val-bercovici-and-callan-fox-weka" ] }, { "slug": "whyhow-ai", "name": "WhyHow.AI", "url": "https://ai.engineer/orgs/whyhow-ai", "summary": "At AI Engineer World's Fair 2025, WhyHow.AI's Tom Smoker described how knowledge graphs, GraphRAG, and multi-agent systems turn documents and public web complaints into structured litigation intelligence. His talk connects these inputs to personalized reporting and case identification for lawyers.", "speakerSlugs": [ "tom-smoker" ], "talkSlugs": [ "knowledge-graphs-in-litigation-agents-tom-smoker-whyhow-ai" ] }, { "slug": "witan-labs", "name": "Witan Labs", "url": "https://ai.engineer/orgs/witan-labs", "summary": "At AI Engineer Europe 2026, Nuno Campos presented Witan Labs' work on financial-spreadsheet agents, combining spreadsheet rendering, a high-fidelity C# engine, and a persistent Node.js REPL. The talk traces improvements through better search, formula tracing, systematic evaluation, and debugging.", "speakerSlugs": [ "nuno-campos" ], "talkSlugs": [ "teaching-coding-agents-to-do-spreadsheets" ] }, { "slug": "workday", "name": "Workday", "url": "https://ai.engineer/orgs/workday", "summary": "At AI Engineer World's Fair 2025, Workday-affiliated Eliza Cabrera joins Freeplay's Jeremy Silva to discuss integrating enterprise AI into core product strategy. The session uses Workday Help and employee self-service to illustrate how AI capabilities can mature within practical enterprise products.", "speakerSlugs": [ "eliza-cabrera" ], "talkSlugs": [ "build-dynamic-products-and-stop-the-ai-sideshow" ] }, { "slug": "you-com-recursive-superintelligence", "name": "You.com / Recursive Superintelligence", "url": "https://ai.engineer/orgs/you-com-recursive-superintelligence", "summary": "At AI Engineer World's Fair 2026, Richard Socher presented an evolution-inspired Eureka machine for automating scientific discovery. His talk connected competing hypotheses and empirical testing with scientific data, simulations, physical experiments, and search infrastructure for AI.", "speakerSlugs": [ "richard-socher" ], "talkSlugs": [ "first-steps-toward-automated-ai-research" ] }, { "slug": "z-ai", "name": "Z.ai", "url": "https://ai.engineer/orgs/z-ai", "summary": "At AI Engineer Code 2025, Yuxuan Zhang presented the open-source GLM ecosystem and GLM-4.6’s coding and preference-benchmark results. The talk connects model development with practical evaluation through CC-Bench, alongside video understanding and resources for accessing and deploying Z.ai models.", "speakerSlugs": [ "yuxuan-zhang" ], "talkSlugs": [ "z-ai-glm-4-6-what-we-learned-from-100-million-open-source-downloads-yuxuan-zhang-z-ai" ] }, { "slug": "zenml", "name": "ZenML", "url": "https://ai.engineer/orgs/zenml", "summary": "At AI Engineer World's Fair 2026, Hamza Tahir presented Kitaru as a durable checkpointing and replay layer for agents. His demonstration addressed what conventional traces leave out: executable runtime state, including the artifacts and environments needed to reproduce an execution.", "speakerSlugs": [ "hamza-tahir" ], "talkSlugs": [ "your-agents-need-a-save-button" ] }, { "slug": "zeta-labs", "name": "Zeta Labs", "url": "https://ai.engineer/orgs/zeta-labs", "summary": "At AI Engineer World's Fair 2024, Fryderyk Wiatrowski and Peter Albert demonstrate Jace scheduling meetings through email. Their talk examines how autonomous agents can take on repetitive work while people retain high-value decisions, and contrasts reactive browser-based agents with limited APIs.", "speakerSlugs": [ "fryderyk-wiatrowski", "peter-albert" ], "talkSlugs": [ "which-jobs-can-be-replaced-today" ] }, { "slug": "zs", "name": "ZS", "url": "https://ai.engineer/orgs/zs", "summary": "At AI Engineer World's Fair 2026, Subbiah Sethuraman and Abhilash Asokan described a pharmaceutical commercial-analytics workflow whose multiple agents divided reasoning responsibility and business context, weakening decisions. Their revised design puts deterministic signal detection and prioritization before execution and gives one agent responsibility for end-to-end reasoning.", "speakerSlugs": [ "subbiah-sethuraman", "abhilash-asokan" ], "talkSlugs": [ "why-we-killed-our-multi-agent-pipeline-subbiah-sethuraman-and-abhilash-asokan-zs-associates" ] } ]