I build and redesign full-stack web apps — with AI where it earns its place.

In finance, edge cases cost real money. That instinct — stress-test before you ship — carried straight into software. Technical skills came from the ground up through real projects, not theory, with every gap in knowledge treated as something to close rather than work around. The foundation matters more than the credential.

The work spans the full stack — frontend through deployment — and the focus is always on what happens after the demo. Architecture that anticipates the API timeout, the double-submit, the malformed response. What breaks matters as much as what works. Every project carries the same commitment: simplicity, clear naming, and code that holds up under review.

Knowing something works isn't the same as knowing why it works. Studying production applications — what was built, what was cut, which tradeoffs shaped the outcome — builds the kind of judgment that tutorials can't. The same rigor applies to AI integration: tools earn their place by solving real problems, not by being trendy. Document every decision. Explain the non-obvious. Code outlives the context that created it.