I build computer-vision and ML systems that actually work.

I'm Etele Kovács, a data science student at the IT University of Copenhagen. I take computer-vision and applied-ML problems from the raw data through to a model that runs and an interface people can use. Every project here is real work, documented down to the metrics and the tradeoffs.

I care most about the unglamorous parts that decide whether a model is actually useful — clean data, honest evaluation, and reproducible pipelines someone else can pick up later. Two of the tools on this site are in daily use by researchers, which is the standard I aim for: not a demo, a thing people rely on.