Hi, I'm Uğur

ML Engineer in Paderborn, Germany

I'm a machine learning engineer with a master's from the University of Tübingen, where I completed my thesis at the Bosch Center for AI. I studied computer engineering at Istanbul Technical University. I like building things and figuring out how they work.

What I Do

I work as a machine learning engineer at Raynet, an enterprise software asset management company, where I built the ML capability from the ground up. That includes fine-tuning and deploying open-source LLMs for automated software identification, building RAG systems and ELT pipelines over millions of software fingerprints, and owning the full MLOps lifecycle from data quality to production.

My longer-term interests are in model development and uncertainty estimation. I write about what I'm learning on my blog.

Research Background

I worked as a research assistant in the Machine Learning in Sustainable Energy Systems group at Tübingen, focusing on probabilistic and explainable forecasting of air pollution using Gaussian Processes, in collaboration with the Eurasia Institute of Earth Sciences at ITU. At ITU Vision Lab, I worked on explainable ML for clinical applications.

Earlier, a research internship at BASIRA Lab led to work on predicting brain connectome changes using graph neural networks, which received the Best Paper Award at MICCAI PRIME 2020. I also spent time as a data scientist at Bahçeci Health Group, building predictive models for infertility treatment outcomes.

Publications

Kaplan, U. A., et al. "Domain-Aware Fine-Tuning of Foundation Models." ICML 2024 Workshop on Foundation Models in the Wild.

Nebli, A., Kaplan, U. A., and Rekik, I. "Deep EvoGraphNet Architecture for Time-Dependent Brain Graph Data Synthesis from a Single Timepoint." MICCAI PRIME 2020. Best Paper Award.

Teaching & Community

I was a tutor for Introduction to Machine Learning at the University of Tübingen. I also worked with inzva, a Turkish tech community, where I developed materials for their Applied AI and NLP study groups and was voted "Best Guide" by participants. I co-organized the VISION Workshop at CVPR 2023.

Outside of Work

I like taking photos of things that catch my eye, mostly street scenes, architecture, and whatever looks interesting while traveling. It's a good excuse to wander around and pay attention.

Get In Touch

Always happy to chat about ML, research, or anything interesting.

GitHub LinkedIn Google Scholar