Divyansh Rai

A universal, persistent memory layer for AI assistants — designed, built, and open-sourced solo.

AI assistants forget everything between sessions. MnemOS solves that by extracting facts from conversations via LLM, embedding them with bge-m3, and injecting relevant context before each new message — invisibly, across ChatGPT and Claude. Memories are stored per-user rather than per-platform, so context follows the person across tools instead of resetting every session.

Research into teaching an agent when to keep a raw memory, compress it to a skill, or crystallize it into a rule.

Most agent memory systems operate at one fixed compression level. AECA studies whether a learned policy can adaptively decide, per memory item, whether to keep it raw, compress it into a reusable skill, or crystallize it into a general rule — and rigorously compares that learned policy against a strong heuristic baseline to find out when reinforcement learning actually earns its cost over simpler methods. Schema, migrations, and the episode-ingestion pipeline are built and tested; the comparison study is in progress.

Research combining test-time weight updates with verifiable-reward training, so a model can learn mid-conversation without absorbing wrong information.

Test-time training lets a model update its own weights from context during inference — but on its own it can't distinguish correct information from incorrect information, so it learns whatever it's told. This project explores pairing test-time training with reinforcement learning from verifiable rewards, so a weight update is only kept when it measurably improves performance on automatically generated, verifiable questions about the conversation. Currently in the foundations and literature-review phase ahead of implementation.