I build practical software solutions: tools, product workflows, automation, integrations, and systems that make complicated work easier to run.
My current focus is CodeRisk.ai.
CodeRisk.ai is an independent review practice for AI-powered products. It looks for hidden trust-boundary, state, consistency, integration, and decision risks that normal tests and happy-path demos often miss.
My work usually sits around:
- building tools and workflows for real software teams
- connecting products to APIs, data, agents, workers, and external systems
- turning ambiguous product problems into shipped, maintainable software
- finding real defects in open-source systems
- turning risk patterns into small, durable fixes
- writing case files that explain why the issue was easy to miss
- improving software behavior at the boundary between users, agents, workers, data, and external systems
I care about software that is useful, maintainable, and grounded in how systems actually behave outside the happy path.