Developer tools and programmable CI/CD
DAgger
Dagger builds software for developers and DevOps teams to define continuous integration and delivery pipelines in code. Its portable DAG execution engine separates pipeline logic from the environment, letting teams run workflows locally or scale them through the cloud. Developers can assemble container environments, cache operations, and trace execution end to end. Dagger also integrates LLMs into software delivery workflows, providing controlled environments for coding agents.
Founded in late 2018 by Solomon Hykes, Sam Alba, and Andrea Luzzardi, Dagger set out to replace the scripts holding software delivery together. Alba and Luzzardi later formed Mendral as a separate company, while Hykes continued building Dagger. Its reusable, composable Functions and modules let teams package and share pipeline logic; the Daggerverse makes public functions searchable. In its January 2025 retrospective, the company reported hundreds of community-built modules and nine SDKs, including community-maintained implementations.
Dagger Cloud supplies commercial observability and module-sharing services, with a free individual tier, paid team plans, and enterprise support. OpenTelemetry-based traces help teams inspect and troubleshoot pipelines. Managed Cloud Engines, in early access, add autoscaling and distributed caching; Cloud Checks, also in early access, connects Git events to checks on that managed infrastructure.
2 talks
Newest firstShip Agents that Ship: A Hands-On Workshop for SWE Agent Builders
Kyle Penfound · Jeremy Adams - Casañas
AI Engineer World's Fair 2025 · 1:21:00
3 speakers at AIE
Affiliations reflect their AIE appearances, not necessarily current employment.
Start here
- Containing Agent Chaos
Start here to understand how model-independent agent workflows can execute across remote infrastructure or CI.
Solomon HykesAI Engineer World's Fair 2025
- Ship Agents that Ship: A Hands-On Workshop for SWE Agent Builders
Follow the workshop for a concrete path from a Hello Dagger template to Claude-assisted changes and a GitHub issue-label-triggered workflow.
Kyle Penfound · Jeremy Adams - CasañasAI Engineer World's Fair 2025
Messages from the stage
Inspecting parallel agent work
Hykes demonstrates Container Use connecting Claude Code through MCP, with asynchronous execution, parallel experiments, and Git-based inspection supporting human intervention.
Balancing autonomy and reliability
Penfound and Adams - Casañas examine narrowly scoped workspace tools alongside TypeScript implementations and nested container execution, grounding the workshop's autonomy discussion in implementation choices.
Affiliations reflect each recorded session, not necessarily current employment.

