I build systems that are meant to run in the real world: reliable, efficient, observable, and worth maintaining.
My background is rooted in Go, backend engineering, observability, and production-grade infrastructure. Recently, I’ve been pushing hard into AI agents — not as toys, but as software systems.
That’s why I’m building maxclaw:
a local-first AI agent in Go, designed to be low-memory, private, UI-ready, and usable out of the box.
I’m especially interested in:
- agent runtimes
- local tool execution
- memory and session architecture
- developer-facing AI products
- the intersection of AI, infra, and UX
I like software with strong bones:
- clear architecture
- boring reliability
- fast feedback loops
- minimal operational drama
If that sounds like your kind of engineering, take a look at maxclaw and say hi.

