Linda Haviv

Linda Haviv is an AI engineer, developer educator, and founder of LindaV Media Labs, helping builders understand AI infrastructure through technical education and working software. A member of the first Nebius Fellows cohort, she brings experience in application development, cloud operations, and developer advocacy to projects that make agent memory and AI discovery easier to investigate.

From television production to cloud engineering

Haviv studied philosophy and worked in television before discovering programming while building a website for a boss. She began exploring code in 2015, took an introductory front-end course, and enrolled full-time at Flatiron School. Returning to FOX as a junior application developer, she moved from front-end work into full-stack JavaScript and Node.js development. A company cloud migration drew her toward AWS; she earned a certification, organized internal learning sessions, and subsequently moved into site reliability engineering. That transition from application development to infrastructure gave her experience both building software and supporting the systems it runs on.

Teaching developed alongside that engineering career. Her 60-second introduction to AWS, published in 2020, translated cloud terminology and infrastructure into a compact lesson originally shared on Instagram. Music and performance also inform her delivery: her public developer profile describes that musical side as “Sing as a Service.”

During three years as a developer advocate at AWS, Haviv helped a small team build and scale its short-form developer education program. This extended her teaching from individual cloud lessons into a broader educational effort: making technical knowledge accessible through brief, focused content. She later worked as a staff developer advocate at Anyscale, educating developers about Ray and distributed AI workloads, before moving into independent building and education. Her account of that transition connects her expanding focus on AI engineering with a teaching practice grounded in hands-on learning. LindaV Media Labs continues that connection between building software and helping others understand how it works.

Making AI infrastructure understandable

Haviv’s projects and teaching give developers concrete ways to examine systems whose behavior can otherwise feel abstract.

• Shared agent memory: Her Second Brain project combines content storage, semantic search, and agent memory in Oracle AI Database 26ai. In-database embeddings make stored material searchable by meaning, and a research agent can query that material. An MCP server exposes the knowledge to different AI assistants, giving developers a working example of how an agent can draw on information beyond a single conversation.

• AI-agent visibility: Her AgentGEOScore project examines website properties relevant to discovery and citation by AI systems. Its reports cover agent access, discoverability, structured data, content clarity, and citation probes, then rank suggested changes. Builders can inspect and test these properties without treating a favorable score as a guarantee that an AI system will cite their website.

• Human-led AI development: Haviv treats coding assistants as help across the development process while keeping developers responsible for planning and decisions. Her approach to AI-assisted development emphasizes reviewing generated code before production use and deciding how an application will reach users before generating it. Infrastructure belongs in that planning: an application that works locally may still fail in its cloud environment. Security scans, monitoring, and observability give developers additional ways to detect problems and assess the software they are shipping.

• Learning through teaching: Haviv encourages developers to explain what they have just learned while they still remember a beginner’s questions. In her discussion of independent building and education, writing and teaching deepen the author’s understanding while helping others make similar transitions. That approach connects her early cloud lessons, her work developing short-form education at AWS, and her newer AI projects: learn the mechanism through building, then make it accessible to another developer.