Agile for Agents Managing Robots The Way We Manage Humans
summary
Agents repeatedly forget decisions and redo work, so Luke builds a minimal Agent Management System (AMS): a directory of markdown files, Agent Handoff Protocol, sprint folders and role prompts (coder, librarian, tester, HR) that create project memory and handoffs. Uses Claude / Claude Code and ChatGPT, integrates with DDEV and a small Go Xdebug helper that runs locally and worked; larger WordPress→Backdrop attempts spawn five failing branches before the handoff+sprint pattern recovers progress.
AI-written summary · about
chapters (7)
- 0:00Introduction and welcome
- 2:13From human to robot management
- 16:30Handoff protocol discovery
- 21:26Building the agent system
- 27:52Live demo of AMS
- 39:18Q&A on agent frameworks
- 48:11Summary and resources
related talks
45mAI agents in practice - beyond the prototype
40mHow To Tame Your Agents?
40mAI Agents in Practice
19mDon't Build Slop (4 Levels of AI Agent Maturity)
52mCut the Hype and Build Your Own AI Agents
27mAgentic Engineering: Working With AI, Not Just Using It
39mAgents for the Sake of Happiness
27mAgentic Coding with Discipline & Skill: Practical Techniques and Tips