Lab

Live demos.

Working examples of the stack I ship. Open one, poke around — if part of it looks like what you need, that's the same stack I'll build for you.

AI Lab

Cloud AI assistant

Chat with an open model via the Gemma API — tokens stream from Google's cloud endpoint. Toggle to a private instance to see the same interface routing to a self-hosted backend.

Gemma APICloud ⟷ Private toggleStreamingOpenAI-compatible

Try the assistant

Cloud by default, private on request.

The demo runs on a cloud API — no infrastructure to spin up on your end. The same interface can route to a private instance on your own hardware. That's the on-premises deployment I offer as a service: a configured, running system handed over, not a setup guide.