Metamind – Cognitive architecture for LLM agents

1 point

6 months ago

Hi HN, I built Metamind because I was frustrated with how stateless modern LLM interactions are. Every conversation starts from scratch, and existing memory solutions felt like overkill for what should be a simple problem.

What is it? A self-contained memory system for AI agents that mimics human memory architecture:

- Episodic memory (conversations, experiences) - Semantic memory (concepts, relations, knowledge graphs) - Vector memory (HNSW-powered similarity search) - Automatic memory consolidation, reflection, and decay

Key technical choices:

- Single Rust binary with single-file database (no Postgres, no Redis, no vector DB) - Multi-user ready with complete data isolation - HNSW for O(log n) vector search instead of naive brute force - Cognitive processes: salience-based decay, memory consolidation, pattern detection

Why share this? I'm genuinely curious:

- Is the "brain-like memory" approach interesting or just over-engineering? - Are others frustrated by stateless LLM interactions? - What memory/retrieval strategies have worked for you? - Where would you take this architecture next?

I know there are limitations, but I wanted to share what's working today rather than wait for perfection. The core retrieval + consolidation loop feels genuinely useful.

Demo video (Skipped to Demo): https://www.loom.com/share/c187767b2d1e4e778bc41676c574408b?t=657