About me

I wrote my first lines of code at 14, building websites and small apps. It was not sophisticated, but it was the first time I realized I could make something out of nothing but logic and patience. A great computer science teacher at school helped turn that spark into a direction.

I started university trying Physics, quickly realized I wanted to build things rather than only theorize about them, and switched to Computer Science at Freie Universität Berlin. That turned out to be the right call. The curriculum pushed me into unfamiliar territory - from operating systems to pattern recognition - and I discovered that I am happiest at the intersection of theory and application.

My master thesis became a turning point. I worked on estimating the direction of traffic sounds using microphone arrays and statistical signal processing. It was my first real taste of translating scientific methods into something that works in the physical world. I have been chasing that feeling ever since.

>From the road to microscopy

After graduating, I joined FERCHAU, an engineering partner of Continental Teves, and started to develop advanced drivers assistance system. However, due to more interesting subjects, I switched to Leica Microsystems, working on image processing for microscopy - 3D visualization, optical distortion correction, and experiments with deep learning for life science applications. It was technically demanding and intellectually rich, and it deepened my conviction that image processing and computer vision is where I belong.

>... and back to the road

In 2019 I moved back to Berlin and into the automotive world, with a short intermezzo in medical imaging. First at Carmeq and then at CARIAD (both Volkswagen Group). There, I worked on bringing machine learning into advanced driver assistance systems, from feasibility studies to production-ready pipelines. The work involved everything from Python prototyping to building abstraction layers for running AI models on embedded hardware. It taught me a lot about the gap between a working model and a shippable product.

>Where I am now

Since 2023, I have been at Sennheiser, working on video processing for unified communication solutions - the kind of systems found in modern meeting rooms. I evaluated embedded platforms, designed video processing algorithms, and shaped how these products are built. It is a role that brings together nearly everything I have learned: image processing, embedded systems, and the pragmatic engineering mindset you need to ship real products.

>Beyond work

I am endlessly curious outside of work too. I run a fairly involved Home Assistant setup, maintain my own server infrastructure (based on k8s), and I utilize my machine learning and MLOps experience also in my personal projects . I regularly dig into topics like edge AI, signal processing, and whatever else catches my attention. I also contribute from time to time to open-source projects like Home Assistant Power Calc. My motto — progress, not perfection — is not just a tagline. It is genuinely how I approach learning and building things.

>Currently learning

>Recently explored

>Links