- more details -

University taught me something: I don't thrive in purely academic setups. Many courses leaned heavily on legacy theory with little connection to current tooling, workflows, or real-world constraints. That disconnect slowly drained my motivation — not because the material was too difficult, but because it drifted away from the reason I'd chosen the field in the first place.

During my Master's studies, this tension became impossible to ignore. My first thesis topic, polyp recognition in gastrointestinal data, had to be dropped due to missing data from the University Hospital Würzburg. It was frustrating, but also my first real lesson in how research can fail for reasons that have nothing to do with competence or effort.

The second topic clicked immediately: Lightweight Interactive Segmentation of 3D Medical and Gastroenterology Data. The work focused on image segmentation: dividing complex medical images into meaningful regions so structures could be analyzed separately. The data ranged from CT and MRI to ultrasound, PET, and microscopic recordings.

Together with a doctoral researcher, I worked hands-on on developing an image segmentation model. We entered the CVPR Foundation Models for Interactive 3D Biomedical Image Segmentation Challenge and placed first. The work culminated in a publication introducing our model, iMedSTAM.

But the long delays caused by the abandoned first topic and the extended writing phase took their toll. Over time, it became clear: my struggle wasn't with Computer Science, it was with the academic format surrounding it. Eventually, I deliberately decided not to complete the Master's thesis.

I don't see that choice as a failure. I see it as clarity. I do my best work when I can build, iterate, and solve concrete problems, not when optimizing footnotes for documents destined to gather digital dust. I bring curiosity and real research experience, and I'm not afraid to question assumptions or call out unproductive complexity when it gets in the way of real impact.

- more details -

You'll often find me hopping between documentation, source code, and small experiments until things click. That process isn't hesitation -- it's how I stay sharp and adaptable. I don't aim to remember everything; I aim to know where to look and how to reason once I get there.

AI helpers are part of that workflow as well. I like using them as sparring partners: for code reviews, exploring alternative approaches, or thinking through a tricky situation from a different angle. They're instrumental -- as long as they don't replace thinking. I prefer to keep my brain fully engaged.

JavaScript remains my home base. TypeScript hasn't been a constant in my work yet, but I'm starting to lean into it.

On the runtime side, I fell hard for Deno and never really looked back at Node.js. The ecosystem feels cleaner, more opinionated, and closer to how I like to work. During my VPNBOX days, I wrote more Bash scripts than I'd care to admit and spent a lot of time configuring GNU/Linux systems and networking. That's the kind of experience that teaches you how systems behave and work.

Containers and virtualization are tools I really enjoy working with. Most of my hands-on experience comes from LXC and LXD, including VM setups with GPU passthrough. I've experimented with Docker, and I'm keen to use it and learn more deeply as the right opportunity arises.

Most of my work so far has been in full-stack web development: building interfaces, wiring APIs, designing UIs, and ensuring the whole system feels coherent rather than fragile. Along the way, I've worked with many tools, sometimes profoundly, sometimes just enough to solve a specific problem.

What matters most to me isn't the length of a tool list. It's knowing how and why to use what's in the toolbox -- and staying curious enough to add new tools when the problem calls for it. And honestly, I'd love to take these habits beyond the web and see where else they apply.

- more details -

I deeply care about free and open source software. Freedom to inspect, change, share, and build on top of things matters to me far more than shiny lock-in. I care about protecting people, not drowning everyone in miles-long legislation that no one fully understands and that hurts consumers and businesses alike.

I'm a harmony-seeking person. A strong team, shared values, and a healthy working atmosphere outweigh hard metrics like salary for me. I do my best work when people enjoy collaborating and when there's trust instead of politics and control.

Outside of work, I move -- a lot. Cycling, rowing, running. Physical activity helps me think, reset, and stay curious. I'm also a big fan of feedback loops, gamification, and systems that reward progress instead of perfection. Small nudges can dramatically change behavior, and I find that endlessly fascinating.

I'm open to traveling for work, and I value flexibility—including the freedom to contribute to open-source projects in my free time or even build something of my own alongside a full-time role. I prefer clarity on these things early so that they won't turn into awkward conversations later.

I care less about grades and more about whether something is purposeful on a deep level. Does it work? Does it help? Does it move something forward? Buzzwords, empty meetings, and slideshows drain my energy quickly. I'm drawn to environments that value usefulness, honesty, learning, and clear thinking.

If you tested me on the spot about some obscure language edge case, I might not have the perfect answer ready. What I do have is a strong mental model of how systems behave, and the ability to reason my way to solutions quickly. I'm unreasonably interested in how small changes in tools, systems, or decisions can have outsized effects -- and I love exploring exactly why that happens.

Curiosity is what keeps me going. And it matters to me far more than any grade ever could.