How I work

I plan before I build. Most of what makes a product work or fail is decided in the architecture, so I spend the time up front on how the pieces fit — the data model, the boundaries between services, what has to be correct versus what can be redone later.

Then I move fast. AI coding tools like Claude Code and Codex changed how much one engineer can ship in a week, and I've leaned into that: I write specs together with the people who need the product, get something running early, and put it in front of them instead of describing it.

The part I care most about is the feedback loop. A team that sees its own product weekly asks better questions than one reading a roadmap, so I try to keep the business close to the tech and shorten the gap between "we should try this" and "it's live."

I like owning things end to end — scrapers and data pipelines, LLM and agent workflows, the web app on top, and the infrastructure underneath. I'm always trying new tools, and I'm happiest on small teams where that's an advantage rather than a process problem.