MemQL exists so AI doesn’t forget what matters.
It’s the memory layer between the models and the applications — the most critical, least-solved part of the stack.
Why it exists
AI has a structural flaw: when a conversation ends, it forgets. So every team rebuilds the same fragmented stack to paper over it — a Postgres next to a vector database next to an event bus next to an LLM wrapper next to a file of retry logic — and keeps it all consistent by hand.
A human brain forgets on purpose. That’s a feature: a brain isn’t built to be an archive, it’s built to think. But when the machine also forgets, the human is the one stuck re-explaining, re-reconstructing, carrying the context the system should have kept.
AI without memory doesn’t amplify you — it hands you back work.
MemQL is the part of the system that remembers. One DSL, on PostgreSQL and TimescaleDB. You describe the behavior; the engine runs it; the memory persists.
What it is
MemQL is an AI-native, time-series memory graph with a single DSL — time-series and event-driven by default, multi-tenant by partition. The full picture lives in the docs and on the homepage; this page is about why it’s built, not a feature tour.
Open source, by conviction
MemQL and MemQL Cockpit are Apache 2.0. No demo, no waitlist — the repo is public and the code is the documentation.
Transparency is total: what you see in the repo is what runs in production. There is no “lite” version behind a paywall and no hidden core. It’s open source now, Apache 2.0, and that’s the commitment going forward. Browse it: znasllc-io/MemQL and znasllc-io/memql-cockpit.
Who’s behind it
MemQL is built and maintained by ZNAS LLC. It’s deliberately code-forward — the work is in the open, and the repository is the resume.
Status
MemQL is Alpha — pre-1.0. It is already running in production against real workloads, but the DSL, the engine API, and the wire surface are still evolving; expect breaking changes between versions. We’d rather say that plainly than pretend otherwise — the honesty is the point.
Get involved
Star it, browse the source, open an issue or a discussion on GitHub. For licensing, enterprise questions, or just to tell us what you’re building, email legal@znas.io. No “book a demo.”