I started coding in low-level systems programming, but everything changed when I built my first autonomous trading agents. Pairing Claude's reasoning with on-chain execution opened my eyes to the power of agentic workflows.
That experience sparked a new obsession: making intelligence run entirely on consumer hardware — no cloud, no API, no surveillance. This led to ClawRS, my most recent project: a blazingly fast Rust runtime for private local inference.
Every project I build answers the same question: "Can this be local, open, and fast?"
I'm deeply invested in the open-source AI movement. I believe the future belongs to open weights, community-driven models, and infrastructure that puts users — not corporations — in control of their intelligence.
