About

I started my career as an editor for a major search engine (back when humans edited the results) before making detours through online advertising technology and product management. While studying fine art at Berlin's Universität der Künste, I worked part-time as a UX writer, localization specialist, and technical author. After graduating, I cut my teeth writing developer documentation and technical tutorials using a docs-as-code approach. I worked as a technical writer for many years while retaining my practice as a multimedia artist and creative coder.

These days, I've slowly crossed over into technical content marketing and am just as comfortable ghostwriting a think piece for a CTO as I am writing a tutorial for a new Python library. Although I've never been a professional developer, I'm no stranger to coding. Through my passion for net and media art, I taught myself to code while working on various art projects — mostly Python, but some Node.js (I haven't entirely given up on learning C#).

Thanks to the machine learning craze, my strongest coding language is Python. After many years working for a software company that dealt in computational linguistics, I became fascinated with natural language processing, and later large language models. Ever since the release of RNNs and later GPT-2 I've been poking and prodding at the limitations of LLMs and have used LLMs in various art projects, such as fine-tuning GPT-2 to generate artist statements for grant applications.

But naturally, my professional writing is all 100% me. Nowadays, my AI tinkering isn't restricted to LLMs, and I'm currently trying to learn ComfyUI and master offline image generation with local models. I'm a big proponent of running your own models, be it Stable Diffusion, LLama-2, or Whisper. I enjoy getting my hands dirty with the underlying technology and tweaking it to my creative requirements.