Senior fullest-stack developer in Berlin. I build systems that have to hold up under real load, and I bet on semantic AI early — generative models, then embeddings, now retrieval. Scroll to follow the path: Berlin → Maastricht → Berlin.
Berlin → Maastricht → Berlin
Berlin
2016–2020 · where it compiled first
A computer-science degree at TU Berlin (final grade 1.8) — but the formative work happened alongside it. At GT-ARC I worked on multi-agent systems; as a student developer at RWTH Aachen I helped build the identity-management apps used by ~47,000 students and 9,000 staff. One habit formed early: build software real institutions actually run on.
En route
2020 · Berlin → Maastricht
In 2020 I left Berlin for Maastricht and an M.Sc. in Artificial Intelligence — a deliberate bet to stop skimming machine learning and go all the way under. The first leg of the loop.
Maastricht
2020–2022 · the bet on generative AI
A master's in Artificial Intelligence (GPA 8.25), all-in on generative models. My thesis trained GANs to generate procedural game content — teaching a network to invent playable levels. This is where the through-line starts: a lattice of weights learning to generate.
Berlin
2023 · 18,000 concurrent players, team of 5
Back in Berlin at Respeak: a chat-based game tied to the TV series Tatort, built before ChatGPT and contracted to survive 100k concurrent. I owned scale — an Azure migration, sentence-embedding models on Azure ML, a Flask refactor that cut redundant SQL and added caching, and load balancing. It launched at 18,000 concurrent with a team of five. The GAN lattice became an embedding space.
Berlin
Respeak · tech lead, building the core
I'm the tech lead on Experte (respeak.io) and built its core: a RAG platform for public administration that returns cited answers grounded in an organisation's own documents (PDFs, SharePoint, Confluence). The through-line closes — GANs → embeddings → retrieval over embeddings.
Berlin
Volunteer dev · the Gesangbuch PWA
Away from paid work, I build for my church — a Gesangbuch PWA (a German hymnal as a progressive web app) the congregation actually uses, in German and English, open source on GitHub. Same engineering care as a client project; the gift is the difference.
frederic@berlin:~$ whoami
systems that scale · AI that ships
The thread runs straight through: a lattice of weights learning to generate, embedding spaces measuring meaning, systems that stay up when the load arrives. I'm still at Respeak, most useful where systems that scale meet AI that ships.
frederic@berlin:~$ contact --open
|
|
|
\|/
V
SCROLL