Sergey Kastryulin - Head Of Applied Research, Yandex Research

Sergey Kastryulin

Head Of Applied Research, Yandex Research

Moscow, Moscow, Russia

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Sergey Kastryulin is a research-driven engineering leader with 9 years of experience building and deploying computer vision and generative models, now heading Applied Research at Yandex Research. He has led diffusion-based text-to-image and super-resolution efforts for YandexART, translating cutting-edge research into production features while keeping computational costs practical. Earlier roles at Philips involved MRI acceleration, network compression and ultra-low-dose CT denoising, reflecting a strong bridge between medical imaging and creative AI. An active open-source contributor, he implemented and refactored key image quality metrics (FID, KID, multi-scale SSIM) in the well-known piq PyTorch repository, improving reproducibility and evaluation for image-to-image tasks. Sergey holds a PhD in AI from Skolkovo and brings multidisciplinary roots in organic chemistry and toxicology to his pragmatic, experimentally rigorous approach to ML.

9 years of coding experience

7 years of employment as a software developer

Upper Intermediate Graduate, General English, IELTS, Upper Intermediate Graduate, General English, IELTS at EF Language Abroad

Doctor of Philosophy - PhD, Artificial Intelligence, Doctor of Philosophy - PhD, Artificial Intelligence at Skolkovo Institute of Science and Technology

Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at Southern Federal University (former Rostov State University)

English, Russian

Stackoverflow

Stats

29reputation

7kreached

3answers

1question

Github Skills (10)

computer-vision10

pytorch10

python10

metric10

generative-model8

xcode6

authentication6

pycharm6

cassandra6

git-submodules6

Programming languages (5)

C#C++SwiftMATLABPython

Github contributions (5)

photosynthesis-team/piq

Dec 2019 - Jan 2023

Measures and metrics for image2image tasks. PyTorch.

Role in this project:

ML Engineer

Contributions:16 releases, 100 reviews, 135 commits in 3 years 1 month

Contributions summary:Sergey implemented and refactored several image quality assessment (IQA) metrics within the `piq` repository, which focuses on image-to-image tasks using PyTorch. They introduced implementations for Frechet Inception Distance (FID), Kernel Inception Distance (KID), and Structural Similarity Index (SSIM), including multi-scale versions and loss functions. The user also refactored and improved the existing SSIM implementation, adding documentation and input parameter validation to the code.

vifbrisquemsepsnrimage-quality

snk4tr/ML-monitoring-helper

Jun 2018 - Dec 2018

Contributions:35 commits, 8 pushes, 5 tags in 6 months

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.

Request Free Trial