Bridging the gap between engineering execution and audio innovation.

I trained as a mechatronics engineer — sensors, actuators, control loops, and a graduation project that put an autonomous guided vehicle inside a hospital where getting it wrong had consequences. That's where the instinct came from: systems fail at the interfaces, so specify the interfaces.

I entered software through QA, which turned out to be the best possible on-ramp to product. Writing 100+ test cases teaches you that a requirement without an observable outcome isn't a requirement. Then I built the delivery pipeline itself — Docker images, GitHub Actions, regression suites — and took 40% off deployment time. Owning the pipe is how I earned the right to decide what goes through it.

Today I run discovery-to-release at Suplyd: 112 tickets authored, Mixpanel funnels as the tiebreaker, and PRDs written so engineers don't have to guess. When something is blocked on a GraphQL response shape or a Docker build, I read the error rather than forward it.

The other half is 13+ years as a musician and producer. Mixing is product work in disguise: you cannot make everything loud. You decide what the listener needs to hear first, cut what competes with it, and ship a balance rather than a wishlist. That's the same muscle as a roadmap.