This is where I keep my notes on building ML systems that stay working.

I’m a slow, deep learner — I’d rather understand one thing from first principles than skim ten. So instead of a static résumé, I keep a working notebook: what I’m building, what broke, and what I learned fixing it.

If you’re a recruiter or engineer in Berlin looking at my projects, this blog is the “why” behind them. If you’re learning MLOps yourself, maybe some of it saves you a wrong turn.

More soon.