ZenSation AI

We build memory for autonomous AI agents.

What is researched here has to survive a Monday morning, not just a review. The ZenBrain Playground runs the open library live in your browser. The paper's results, their method and the correction history are at zensation.ai/en/forschung/benchmarks.

ZenBrain is a seven-layer, neuroscience-derived architecture for agent memory in LLM agents (paper version 9, arXiv:2604.23878v4).

🧠 What we build

• 7-Layer Memory Architecture — working, short-term, episodic, semantic, procedural, core and cross-context memory; five correspond to established cognitive constructs, core and cross-context are engineering layers
• Self-hostable — runs on your own infrastructure: no external processor, no third-country transfer
• Open Source — code under Apache-2.0, research records under CC BY 4.0 or Apache-2.0
• Made in Germany 🇩🇪

🚀 Featured

• 📄 Paper: arXiv 2604.23878 — ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture
• 🤖 Model: zensation-ai/zenbrain
• 📚 Collection: ZenBrain: agent memory research
• 🆔 Research records: open records on Zenodo, each with version and concept DOI — full list

🔗 Links

• 🌐 Website: zensation.ai
• 💻 GitHub: github.com/zensation-ai
• 💼 LinkedIn: zensation-ai
• 🆔 ORCID: 0009-0001-1793-012X
• 🎓 Google Scholar: Alexander Bering

👤 Founder

Alexander Bering — Division Manager in a mid-sized company and Principal Investigator. Founder @ ZenSation.