About

I started in electrical engineering and left it for cognitive science — the interdisciplinary study of cognition, from neural networks to human reasoning. That decision shaped everything since: I approach AI from both sides, the mathematics of how models learn and the cognitive science of why they behave the way they do.

At the University of Osnabrück I focused on deep learning, reinforcement learning, and probabilistic modeling, finishing with a thesis on computational belief modeling (grade 1.0, Musslick Lab). Alongside the degree I led projects at a US-registered startup, built and ran production software for German SMBs, and worked industrial jobs that paid for all of it.

Today I work as the first data & AI hire at a German architectural glazing manufacturer, building their agentic lead-generation infrastructure — and I co-run PredOpt, a small consultancy for AI-driven process optimization in DACH logistics.

I work in German (C1), English, French, and Arabic.