Summary
Arthur Fedorov is a machine-learning and econometrics specialist with 8 years of experience building production ML systems and leading teams at Yandex.Self-driving. He combines deep theoretical work—manifold learning on time series with DTW- and KL-based metrics, stochastic colored Petri nets for transport flows—with practical engineering: BERT-like models, distillation for runtime, crowdsourced labeling, and end-to-end pipelines. Previously at Sberbank he delivered high-impact CLTV and churn models that translated to multi‑hundred‑million‑ruble gains and mentored many junior data scientists to rapid promotion. Comfortable across computer vision, NLP, simulation and causal inference, he pairs quantitative research from his PhD work at RANEPA with product-focused delivery. Based in Odintsovo, he is notable for translating advanced information-geometry ideas into faster, business-facing ML workflows.
8 years of coding experience
3 years of employment as a software developer
Bachelor's degree Quality Control Technology/Technician, Bachelor's degree Quality Control Technology/Technician at Moscow State University of Transport (MIIT)
PhD student Economic Theory, PhD student Economic Theory at The Russian Presidential Academy of National Economy and Public Administration (RANEPA)
Russian, English