I build applied and agentic AI systems, with a particular interest in making them intuitive, intelligent, auditable and safe to give real permissions to. Previously: AI governance at the European Commission, Go/Kubernetes engineering at SAP, and product management at Carealytix. Currently finishing an MSc at TU Dresden. Below: selected work, and occasional notes.
Recommended articles
Deleting the most beautiful thing I builtA procedurally rendered sky drove the whole interface for months: sun position, clouds, glassmorphism, all of it. It looked best on the machine I built it on and worst on the one I actually use, so it's gone.Building a search that admits when it found nothingA journal is a bad corpus to search: bilingual, untitled, no shared vocabulary. Three retrieval arms, and a cut that is allowed to return zero results.Capture first: Building a journal with no filing stepA journal that never makes you file a thought before you've finished having it. Fully private and self-hosted.
All writing →