GEPA: The Self-Improving AI Agent Loop

Watch how Pydantic AI agents continuously optimize in production — from traces to evals to auto-optimized prompts.

Press Start to begin the loop
The GEPA loop shows how live AI agents self-improve using traces, evals, and prompt optimization.
Pydantic AI Logfire GEPA Evals Feedback Loops
Enjoy this tool? Build your own with Super