Ship Agents that Ship: A Hands-On Workshop
summary
Builds a sandboxed, production-minded AI developer agent using a Dagger workspace module plus real LLMs (Claude/GPT), languages (Python/Go/TS), and containerized test runtimes; the agent is a callable develop function agent that lists/reads/writes files, runs the project’s test step inside containers, and returns a completed directory. Workflow runs locally or as a GitHub Actions job triggered by an issue label, producing an actual PR that closes the issue and showing execution in Dagger Cloud.
AI-written summary · about
chapters (8)
- 0:00Introduction and setup
- 5:31What is Dagger?
- 10:08Dagger basics
- 12:48Constructing the pipeline
- 21:15Creating the agent workspace
- 28:20Building the AI agent
- 42:28Configuring the agent
- 55:13Automating with GitHub
related talks
15mAgents Don’t Ship Themselves
11mBuilding Secure Backend Services with AI Agents
31mReal-World Multi-Agent Development | Ship Faster with AI Agents
30mBuilding Tools with Tools: Modern AI-Driven Development
52mCut the Hype and Build Your Own AI Agents
40mHow To Tame Your Agents?
40mAgents in Action: Advancing Open Source AI Missions
30mMind the Gap - A Developer's Roadmap To Building AI Agents