Real memory.A cognitive architecture.Collective intelligence.
Humans and machines on an architecture that retains knowledge and structures decisions — researched in the open: from AI-assisted knowledge work to safeguarding public spaces, without biometric identification.
ZenBrain is our open-source memory core. Multiple applications build on it — the same seven-layer architecture, only the domain adaptation varies.
Memory layers
01Working
02Short-Term
03Episodic
04Semantic
05Procedural
06Core
07Cross-Context
The architecture is designed domain-agnostically. Research partnerships, public-sector cooperations, and industrial applications can build on this open-source core without modifying it.ZenAI in production: the same ZenBrain architecture as an AI operating system for knowledge work — the same approach extends to further domains.
Research on early recognition of critical situations in public spaces and at mass-casualty incidents — explicitly without biometric identification, with civil liberties built into the architecture by design.
What we explicitly do not do
✗No facial recognition
✗No biometric identification
✗No real-time person tracking
What we work with
✓Skeletal and motion data — without facial features
✓Three-outlier model differentiating source / helper / knower
✓Helper protection as an explicit objective
Research prototype · GDPR Art. 89 · designed under the EU AI Act
Empirical results from the open-access preprint (arXiv 2604.23878), LongMemEval-500 benchmark with three independent LLM judges. Methodology and full results on the publications page.
Proof, not claims
From paper to production
Our cognitive architecture runs in production systems today. The same algorithms we publish power real software — evidence that the research carries through.
✓
ZenAI — The AI operating system for the enterprise — consolidates data streams, self-hosted, running in production
✓
ZenCRM — A shipped iOS app built on the ZenBrain architecture
