Twenty-two years ago I wrote Java. Today I build AI agent systems.

I started as a Software Engineer writing Java and J2EE applications fresh out of university. Within a year I was deep in PLM and Enovia consultancy — a niche that took me across some of the most complex engineering environments in the world: Texas Instruments, NCR Corporation, IBM, and eventually Adidas through Infosys.

The pivot to Product came naturally. After nearly a decade as a PLM technical expert, I moved into Technical Product Owner and then Senior Product Manager roles at GfK SE in Nürnberg — spending six years driving legacy transformation across Retail, Hi-Tech, and Consumer Goods analytics at one of Germany's most respected market intelligence companies.

In 2026 I stopped reading about AI and started building with it. The catalyst was simple: I had real problems — property documents, German tax filings, investment monitoring — eating my weekends. I decided the best way to understand AI's real capabilities was to apply it to problems I already knew intimately. What I built and what I learned is what I document here.