Sergii Dymchenko
Research Engineer at Meta
Bellevue, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolSergii Dymchenko is a seasoned research engineer with 15 years of experience building robust software systems and shipping production-grade ML infrastructure. Based in Bellevue, he blends deep interests in algorithms, programming languages, and AI with a pragmatic drive to use the right tool for the job. At Meta and previously Google, he has focused on modernizing complex build and deployment pipelines and improving developer workflows. An active contributor to the PyTorch ecosystem, Sergii has streamlined CI/CD, removed legacy build cruft, and fixed tutorial and vision library issues—work that directly supports one of the most widely used deep learning frameworks. His background ranges from hands-on Python development and community-facing roles at TopCoder to infrastructure engineering, revealing a rare mix of systems-level thinking and attention to developer ergonomics.
15 years of coding experience
7 years of employment as a software developer
Computer Science, Computer Science at Donec'kij Nacional'nij Tehnicnij UniversitetBSc, Computer Science, BSc, Computer Science at National Academy of Management, Kyiv
English, Russian, Ukrainian
Stackoverflow
7,039reputation
711kreached
234answers
12questions
prolog
top-5%
pytorch
top-5%
python
top-5%
Github Skills (26)
pytorch10
python10
machine-learning10
build-system10
cicd10
automation10
computer-vision10
automations10
documentation10
dockers9
pytest9
docker9
resource-loading9
prolog9
data-loading9
Programming languages (14)
Github contributions (5)
pytorch/pytorch
Aug 2018 - Jan 2023
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
DevOps Engineer
DevOps EngineerContributions:1 release, 1298 reviews, 342 commits in 4 years 6 months
Contributions summary:Sergii primarily focused on improving the build and deployment processes. Their commits involved removing outdated configurations, such as hardcoded flags and legacy build environments (e.g., xenial gcc5.4), while also migrating off older build systems (e.g., circleci) in favor of updated ones. They also addressed various other system related changes. This indicates a focus on streamlining and modernizing the repository's build and testing infrastructure.
pythongpu-accelerationdeep-learninggpunumpy
pytorch/tutorials
May 2023 - Jan 2025
PyTorch tutorials.
Role in this project:ML Engineer
ML EngineerContributions:35 reviews, 13 PRs, 3 pushes in 1 year 8 months
Contributions summary:Sergii contributed to the PyTorch tutorials repository by fixing bugs, updating deprecated APIs, and correcting formatting issues. Their work includes addressing issues related to the `docathon-label-sync.py` script, updating the TorchVision pretrained API, and correcting a typo. The commits also show the user updating code related to `torch.symeig` and the `torch_compile_tutorial.py` file.
deep-learningpytorchpytorch-tutorials
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.