About

I’m Nauryzbay, a machine learning engineer at Lightly in Zürich. Computer vision is the thread running through most of what I’ve worked on, and I’m especially drawn to the part where a promising research result has to survive contact with real data and a latency budget.

Before Lightly I finished an M.Sc. in Data Science at ETH Zürich and spent a year at the AIT Lab — first a master’s thesis on generating clothing geometry from a single image with diffusion models, then a research assistantship working on a motion capture pipeline and body registration.

Earlier: research internships at Sony R&D in Stuttgart and Zürich on face capture with event-based vision sensors; a stint at Plask in Seoul getting a pose and shape inference pipeline to run in real time, which meant TensorRT, half-precision, and rewriting the thing in C++; and research at KAIST’s Visual AI group on guiding RANSAC with a neural network to fit geometric primitives to point clouds. My bachelor’s is in EE and CS, also from KAIST.

The best way to reach me is LinkedIn. My code lives on GitHub.