I've been building software for a few decades now. Most recently, I founded LaunchPT, an AI consultancy focused on agent architecture and evaluation frameworks. Before that, I led Explainable AI/GenEval at Credit Karma, shipping production GenAI systems with LLM-as-a-Judge evaluation and policy-driven guardrails across GPT, Sonnet, and Mistral models.

Earlier, I led infrastruture engineering for FDA compliance for clinical trial platforms at Medidata and FAA compliance for aviation maintenance systems supporting the US Navy, KLM, Southwest, Qantas, and Boeing's Dreamliner.

I've keynoted and facilitated workshops at conferences across four continents—TestBash, Agile Testing Days, Let's Test, European Testing Conference—and served as Program Chair for the CI/CD track at the Agile Alliance Conference.

My current research explores Strategic Reset Architecture: giving AI agents what humans struggle with—the ability to doubt early assumptions and start over with accumulated wisdom. We've spent decades learning that complex problems deserve humility. Our agents should learn that too.

Based in Berlin. It's a really fun time to be in tech :)