I’m Julius, a software engineer working at the intersection of backend architecture and applied AI. I like making complicated things feel simple.
Explore my workTools that bring useful context into engineering work. A custom coding assistant with retrieval, automated knowledge updates, and Azure-hosted language models, alongside an LLM-supported innovation toolset.
As the solo developer, I took a cultural app from architecture to release: a .NET backend, SQL database, administration tools, and Flutter apps for iOS and Android. The product received the Urbanana Award, the Dortmund Heimatpreis, and the Cultural and Creative Pilot title.
A vision transformer that predicts a Counter-Strike 2 player’s location from an image. I built a dataset of 1.7 million images and trained the model on an RTX 4080. On unseen data, its median error was roughly three in-game footsteps.
I’ve been building software since 2013, from client projects at netzlab to a complete cultural app, and now engineering tools at Wilo. My focus is on clear architecture, thoughtful implementation, and technology that earns its place.
I hold an M.Sc. in Computer Science from TU Dortmund. Outside software, I’m into photography, learning languages, and finding better ways to organize information.