Oleg Filatov
AI Researcher at Helmholtz AI
Cologne, North Rhine-Westphalia, Germany
Summary
👤
Senior🎓
Top SchoolOleg Filatov is an AI researcher and senior machine learning engineer with nine years of experience building ML frameworks and production-grade models at top research labs including Helmholtz AI, CERN, and DESY. He combines deep academic training (MS and PhD work in machine learning and physics) with hands-on systems work—contributing to large-scale scientific software like the CMS Offline Software where he implemented and validated the DeepTau ID algorithm. Comfortable across C++, Python, and HPC environments, he focuses on turning complex physics-driven ML research into robust, testable pipelines on supercomputing infrastructure. Based in Cologne, he bridges research and engineering, pushing large language model and ML-tooling advances while retaining a strong emphasis on reproducibility and integration into established scientific stacks.
9 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy - PhD, Machine Learning / Physics, magna cum laude, Doctor of Philosophy - PhD, Machine Learning / Physics, magna cum laude at University of HamburgBachelor of Science - BS, Applied Physics and Mathematics, 8.8/10.0 (with honours), Bachelor of Science - BS, Applied Physics and Mathematics, 8.8/10.0 (with honours) at Moscow Institute of Physics and Technology (State University) (MIPT)
English, Russian, German
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (13)
algorithm10
algorithms10
cms10
offline-web-app10
high-energy-physics10
deep-learning10
c-language10
cprogramming-language10
offline-app10
tensorflow9
machine-learning9
web-framework8
python8
Programming languages (4)
Github contributions (5)
cms-sw/cmssw
Apr 2022 - Jul 2022
CMS Offline Software
Role in this project:
Back-end Developer
Back-end DeveloperContributions:50 commits in 2 months
Contributions summary:Oleg primarily contributed to the development of the DeepTau ID within the CMS Offline Software repository. Their work focused on implementing and refining the DeepTau algorithm, specifically version 2.5. This involved setting up configuration files, modifying core C++ code for the DeepTau ID, and adjusting the scaling parameters used by the model. The user also made changes to testing scripts, demonstrating a focus on integrating and validating the DeepTau model.
cmscernweb-appc-plus-plusbackbonejs
ofivite/tatoo
Jul 2022 - Oct 2022
Tau Toolkit: a framework with a TensorFlow implementation of graph/particle convolution- and attention-based models for tau lepton identification.
Contributions:18 reviews, 4 PRs, 60 pushes in 3 months
particlepythontau-leptonconvolutiontau
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.