Shodh Memory

Shodh Memory is a cognitive memory system for AI agents inspired by neuroscience, implemented as a lightweight (~30MB) Rust binary. Unlike traditional vector databases or LLM-based memory management, it uses Hebbian learning rules and decay curves for storing, strengthening, and fading memories, resulting in sub-millisecond, fully offline operation. It features a knowledge graph architecture, hybrid ranking, and Model Context Protocol (MCP) support, making it ideal for robotics, IoT, and edge computing environments requiring privacy and real-time performance.