FindingNEMO
Built at The Recursive Agents Hackathon · Oct 17, 2026 · San Francisco, CA
LLM agents learn can learn skills from action sequences that worked before. But a skill doesn't know why it worked, so when the world changes, the agent replays it and fails again and again. FindingNEMO learns verified rules in training. When the world contradicts one, it notices at the exact step, revises with a single LLM call, and recovers in the same episode, where a skill-only agent keeps repeating the mistake.
