Andrey Talman
Software Engineer at Meta
Montreal, Quebec, Canada
Summary
🤩
Rockstar🎓
Top SchoolAndrey Talman is a seasoned software engineer with 10+ years of experience building and automating large-scale systems across industries from international organizations to FAANG. Currently at Meta, he combines back-end engineering and DevOps expertise—demonstrated by substantial contributions to the PyTorch ecosystem—optimizing CI/CD, cross-platform builds, and dependency management for complex ML projects. His foundation in C++, C#, and modern JavaScript (Node, ES6, React) lets him bridge low-level performance work and web-facing interfaces, while past roles show leadership in delivery and architecture. Comfortable in Agile/Scrum environments, he has a track record of shipping reliable build pipelines (including manylinux, CUDA, and macOS M1 support) that quietly keep ML tooling usable for thousands of developers.
10 years of coding experience
18 years of employment as a software developer
Bachelor of Applied Science (BASc), Computer Science, Bachelor of Applied Science (BASc), Computer Science at McGill University
Github Skills (32)
github-ci10
javascript10
docker10
pytorch10
python10
machine-learning10
bash10
build-system10
conda10
dockers10
cicd10
automation10
ci-cd-pipeline10
build-automation10
cuda10
Programming languages (11)
Github contributions (5)
pytorch/builder
Nov 2021 - Jan 2023
Continuous builder and binary build scripts for pytorch
Role in this project:
DevOps Engineer & Automation Engineer
DevOps Engineer & Automation EngineerContributions:290 reviews, 179 commits, 1008 PRs in 1 year 2 months
Contributions summary:Andrey primarily contributed to automating and maintaining the build process for the PyTorch project. Their work focused on updating build scripts, Docker configurations, and CI/CD pipelines to support new CUDA versions, Python versions, and operating systems, specifically Windows. The user implemented and refined scripts for various build types including conda, wheel, and libtorch, demonstrating a broad understanding of the project's build infrastructure and dependency management. Furthermore, the user also made changes related to smoke testing and test execution, reinforcing their role in build validation.
pytorchbuild-scriptscontinuousbuilder
pytorch/pytorch
Nov 2021 - Jan 2023
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:8 releases, 1585 reviews, 271 commits in 1 year 2 months
Contributions summary:Andrey's contributions primarily involved modifying the build system and CI/CD pipeline of the PyTorch repository. They focused on addressing build errors related to dependencies and the Docker build process, including issues with the manylinux build environment and the installation of Python dependencies. The user implemented solutions by modifying the build scripts, updating versions, and configuring the environment to ensure successful builds, ultimately contributing to the stability and maintainability of the PyTorch build process.
pythongpu-accelerationdeep-learninggpunumpy
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.