Summary
Mykyta Shchedrolosiev is a software engineer and PhD physicist based in Hamburg with eight years of experience building scalable machine learning pipelines for large scientific datasets. He designed and deployed CNN and GNN models for particle identification and rare-event detection at DESY/CMS, improving classification and simulation-to-data agreement while leading a team and contributing production-ready tools to international collaborations. Comfortable across data ingestion, domain adaptation, hypothesis testing and end-to-end evaluation, he has handled workflows for 10+ TB datasets and translated research algorithms into robust analysis software. Earlier work includes GEANT4-based detector simulations and custom analysis frameworks for collider data, evidencing deep domain expertise in high-energy physics. Now at freiheit.com technologies, he brings research-grade rigor to engineering problems, pairing experimental intuition with practical software delivery. A lesser-known strength is his track record of optimizing loss functions and training objectives (e.g., Asimov loss experiments) to directly improve physics-analysis performance.
8 years of coding experience
Doctor of Science Physics, Doctor of Science Physics at University of Hamburg
Master's degree (Physicist junior researcher) High Energy Physics, Master's degree (Physicist junior researcher) High Energy Physics at Taras Shevchenko National University of Kyiv
Ukrainian, English, German, Russian