Sasha Sidorov
Software Engineer at Meta
California, United States
Summary
🤩
Rockstar🎓
Top SchoolSasha Sidorov is a seasoned ML systems engineer with 12 years of experience building and scaling machine learning infrastructure for top-tier tech companies, currently shepherding LLaMA deployments at Meta. He has led foundational ML infra initiatives at Cruise—driving PyTorch transitions, monorepo adoption, orchestration, lineage and parameter management systems—and previously helped create Facebook’s Caffe2/PyTorch ecosystem where his work on RNNs and fleet-wide optimization produced material capacity savings. Comfortable moving between low-level performance wins (10x data-read speedups early in his career) and org-level technical strategy, Sasha combines hands-on kernel and operator development with mentoring and culture-building. An active open-source contributor, he added debugging utilities and core operators to the famous Caffe2 repo, reflecting a pragmatic focus on developer ergonomics as well as production robustness. Based in California, he pairs academic training from Saint Petersburg State University with a track record of launching high-impact, cross-functional ML platforms.
12 years of coding experience
11 years of employment as a software developer
Master's degree Computer Science, Master's degree Computer Science at Saint-Petersburg State University
English, Russian
Github Skills (9)
rnn-model10
caffe10
deep-learning10
n10
python9
ai9
machine-learning9
debug9
debugging9
Programming languages (3)
Github contributions (5)
facebookarchive/caffe2
Jan 2017 - Mar 2018
Caffe2 is a lightweight, modular, and scalable deep learning framework.
Role in this project:
ML Engineer
ML EngineerContributions:156 commits, 16 PRs, 5 pushes in 1 year 2 months
Contributions summary:Sasha primarily contributed to the Caffe2 deep learning framework, focusing on debugging, and improving its usability for developers. They introduced a `DebugMode` helper class and a `@debug` decorator to simplify debugging workflows. The user also implemented the `FCTransposed` operator and worked on the integration of Static RNNs.
pytorchscalablecaffe2deep-learningml
salexspb/pytorch
Sep 2018 - Sep 2019
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:2659 pushes, 22 branches in 11 months
pythongpu-accelerationdeep-learninggpuacceleration
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.