I’m German and live near Munich. I came to ML through electrical work, automotive engineering, and embedded systems.
I’m interested in learning how systems behave from messy data, figuring out where simulation misses reality, and getting models running on hardware.
Outside work, I’ve spent long stretches traveling across Asia and Australia.
Why physical systems
I got into batteries somewhat by accident, working on an 800 V prototype.
That’s still the kind of engineering I enjoy: starting with a rough idea, working it through from theory to hardware, and testing what actually holds up — or where my assumptions were wrong.
How I like to work
I learn by building and going down the stack. I like owning problems end-to-end, making decisions, and having them challenged to find the best trade-offs.
I like simple, inspectable systems. Removing a line from my dotfiles gives me more joy than adding the tool of the week.
For ML, that means making inspection easy enough that I actually do it. I start with the simplest approach that could work, then test it on the actual task.
Selected work
- batgrad: An end-to-end system for neural modeling of battery dynamics.
- Master’s thesis: Benchmarked neural models for battery state estimation.
- 800 V automotive battery prototype: Owned electrical-interface design through integration and validation.
- Smart-glove prototype: Built sensor electronics, FPGA acquisition, and real-time ML classification.