Build Production Ready AI Agents with Burr
summary
Burr is an open-source OS library that claims to make production AI agents reliable by treating apps as graphs of actions, conditions, and central state and by providing concrete tooling: sqlite persistor, open-telemetry integration, halt-before human-in-the-loop, fork and replay. Code examples show one-line persisters (sqlite/postgres/mongo/redis), async step/stream APIs, pydantic typing plugins, LangChain/hStack/Hamilton adapters, a UI that records runs and lets teams annotate and export test sets. Demos and early customers run, debug, fork, and iterate on real failures.
AI-written summary · about
chapters (13)
- 0:00Building AI agents with Burr
- 0:48Speaker background and Burr overview
- 2:55AI agents and challenges
- 5:40Pitfalls in AI agent dev
- 7:46AI agent workflow
- 9:54Burr’s graph model
- 14:06Human-in-the-loop and persistence
- 16:33Monitoring with Burr UI
- 18:38Debugging in production
- 20:22Burr integrations
- 22:29Why Burr simplifies dev
- 24:53Roadmap and features
- 25:58Q&A
related talks
34mBuilding Ai Agents with Shell Scripts
46mFullstack AI: The Building Blocks for Production Ready AI-Powered Software
27mHow to Build AI Agents and Keep Your Sanity
17mHow to Make Useful AI Agents
39mTowards Useful, Reliable, and Scalable AI Agents- 39mTowards Useful, Reliable, and Scalable AI Agents
52mCut the Hype and Build Your Own AI Agents
28mBuilding and Sharing Reliable AI Agents with BeeAI