cv

Here is my CV. You can download a pdf version at the right.

Basics

Name Egor Petrov
Label Machine Learning Researcher
Email petrov.egor.d@phystech.edu
Url https://moderntalker.github.io
Summary Machine Learning Research Resident at Yandex Research. Research on efficient deep learning, scaling laws, optimization, and large-scale training systems.

Work

  • 2025.07 - Present
    ML Researcher
    Yandex Research
    ML Research Residency, Moscow
    • Spearheading research on asynchronous pipeline parallelism for efficient large-scale model training.
    • Designing and analyzing novel algorithms for highly distributed systems to enhance scalability and performance.
  • 2024.08 - 2025.01
    ML Engineer
    Yandex
    Personalization Quality R&D Group, Moscow
    • Engineered a production pipeline integrating Vowpal Wabbit-based online features into a large-scale CatBoost model, achieving a 0.1% uplift in core ranking metrics (AUC, nDCG).
    • Implemented a parallelized, MapReduce-style data converter that accelerated data processing by 20x and reduced model training time from hours to minutes.
  • 2024.02 - Present
    ML Researcher
    Laboratory of Mathematical Methods for Optimization, MIPT
    Moscow
    • Pioneered memory-efficient zeroth-order optimization methods for fine-tuning LLMs, achieving a 50% reduction in memory footprint.
    • Developed and theoretically analyzed novel stochastic and zeroth-order algorithms for decentralized optimization, validated on variety of domains.

Education

  • 2025.01 - Present

    Moscow

    Yandex School of Data Analysis
    Machine Learning
  • 2024.09 - 2026.06

    Moscow

    Data Science Track (MIPT & Yandex School of Data Analysis)
    Specialized curriculum in ML, DL, RL, CV, NLP, RecSys, and Time Series Analysis
  • 2023.09 - 2024.06

    Moscow

    Deep Learning School
    Neural Networks and Computer Vision
  • 2022.09 - 2026.06

    Moscow

    B.S.
    Moscow Institute of Physics and Technology (MIPT)
    Applied Mathematics and Informatics

Publications

Interests

Efficient Deep Learning
Optimization
Theory of Learning
Scaling Laws

Projects

  • - Present
    Lead Developer, ZO-Library
    Engineered a comprehensive open-source library for zeroth-order (ZO) optimization, implementing a wide range of state-of-the-art algorithms. Designed with a user-friendly, torch.optim-style API for seamless integration into existing ML pipelines.
    • GitHub & Publication Forthcoming
  • 2024.01 - Present
    Teaching Assistant & Mentor
    Mentored undergraduate students in Algorithms, Data Structures (MIPT), and Introduction to AI (Central University), leading project-based learning and providing technical guidance.
    • 2024 – Present