Our moat is a technology base we build ourselves
Technology stack
Behind the consulting stand our own products and libraries: Balansis — an open numerical library, MagicBrain — open-source neural research, KnowledgeBaseAI — the knowledge layer, xteam-agents — agent orchestration, Agent Native Universe — an open runtime where agents interact reproducibly. StudyNinja runs in production together with the knowledge layer (closed beta), and AVA works out the client’s economics. Every maturity status is an honest one.
Active Development
Balansis
Numerical library · PyPI 1.1.0 · Lean 4
A computation library in which division by zero, overflow and loss of precision do not crash the calculation but become explicit and auditable. The algebra of the model is proved in Lean 4; the code is open on PyPI (1.1.0, AGPL-3.0 or a commercial licence).
Research
MagicBrain
Neural core · Research
A research neural layer: spiking networks, local learning and a network architecture specified by a compact genome rather than designed by hand. Our own line of research, not a product: the code is open (PyPI 0.8.0) and is not used in products yet.
Active Development
Agent Native Universe
Agent runtime · MIT · reproducible
An open-source (MIT) runtime for self-organising agents: relationship negotiation, a resource economy, Byzantine agreement between nodes and recursive organisation into groups.
Active Development
KnowledgeBaseAI
Knowledge platform · graph + semantic search
An engine that turns documents and a domain into a connected model: entities, relations, semantic search and knowledge provenance. A separate, reusable layer.
Active Development
xteam-agents
Agent orchestration · YAML → LangGraph
Assembling multi-agent automation from YAML specifications without rewriting code: the specification produces the graph and the agent topology.
Production
StudyNinja
Learning platform · in production · closed beta
A preparation platform for school students: a subject knowledge graph, gap diagnostics and selection of the next learning step. It runs in production — so far at closed-beta scale, not as a demo.
Cost the effect up front
Want to apply this to your task?
Describe it in your own words — we will work out which combination of technologies solves it and whether it pays off, before any work starts.
Write to us