Agents need memory, not just access.
Real-time data co-located with agent runtime, not API calls to stale exports. The twin is the context window — always current, always verified.
About
Built for the networks AI will run on.
Mission
Why OmniTwin exists.
Network infrastructure management hasn't kept pace with the complexity it serves. Teams still reconcile spreadsheets against CLI output. Drift between intended state and actual state goes undetected until something breaks.
OmniTwin was built to close that gap — a network twin that understands topology, tracks drift in real time, and lets AI agents operate on a shared, verified model of the network.
The team
Kortesa Labs
We're a small team of infrastructure engineers, distributed systems builders, and AI practitioners. We've operated networks at scale, built developer tools, and shipped production ML systems. OmniTwin is where those disciplines converge.
Principles
What we believe.
Two databases are better than one.
Intent (Postgres) and reality (Neo4j) are structurally different. Forcing both into one schema hides drift. We keep them separate on purpose.
Open standards, not lock-in.
REST, gRPC, MCP, SSE. Every protocol your stack already speaks. No proprietary SDK required to get value on day one.
Deployment flexibility is non-negotiable.
From a single binary on a laptop to air-gapped on-premises. Same platform, same API, same agents — regardless of where it runs.
Ship the math, not the UI.
The hard problems are graph reconciliation and drift detection, not dashboards. The interface is a command bar, not a control panel.
Open source
Standing on open shoulders.
OmniTwin is built on open foundations: Postgres, Neo4j, the Go and Rust ecosystems. These projects represent decades of collective engineering work, and we take that seriously.
As the platform matures, we plan to contribute tooling back — particularly around graph reconciliation, CIDR arithmetic libraries, and agent protocol definitions. The infrastructure community gave us the building blocks. We intend to return useful primitives.
Postgres 17 + pgvector · Neo4j · Go · Rust · gRPC · MCP
Private Beta
Build with us.
We're onboarding the first teams now. If your network is where AI will operate, we'd like to talk.
Join the beta