recursive-auto-improvement

Introduces the Pro Agent Builders series, focusing on recursive auto-improvement (RAI) where a coding agent runs hundreds of probes overnight to automatically improve another agent's performance.

The author demonstrates how to build a recursive auto-improvement loop for AI agents: a coding agent mines usage data to create probes, runs them against a live agent, and updates the agent's instructions and tools until all probes pass. The walkthrough uses the Agno/AgentOS platform and the Radar agent as an example.