ust open-sourced Butterfly RSI - a recursive self-improvement framework that gives LLMs actual memory and personality evolution

Tested across multiple models. Implements mirror loops + dream consolidation so AI can learn from feedback and maintain consistent behavior.

Built it solo while recovering from a transplant. Now looking for collaborators or opportunities in AI agent/memory systems.

Check it out: https://github.com/ButterflyRSI/Butterfly-RSI