ai_brain

Self-hosted semantic search over the lab’s own knowledge base — pgvector, llama.cpp embeddings, hybrid keyword+vector retrieval, and a warm HTTP service for sub-second recall. Not a chatbot memory gimmick — infrastructure the lab’s own AI agents run on. A memory-provider plugin extending it to our Hermes agent deployments is in design, not yet shipped.

What it is, plainly

A frontier application built for a Malaysian constraint that existing tools either ignore or addressed last decade. The lab built it because no one else had — and could back it with a build log, not a marketing deck.

Why this exists

ai_brain isn’t a product for a customer — it’s part of the tooling the lab’s own work depends on. It’s documented here because the lab publishes what it builds, not because it’s for sale.

Where it stands

Status: Production.
This is an honest label. No “coming soon” theatre, no inflated readiness claims — if it’s experimental, we say so. If you want the current build log, talk to the lab.