Peilin Guo
Software Engineer at Google
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolPeilin Guo is a software engineer based in Mountain View with five years of experience building and optimizing ML and infrastructure tooling, currently on the engineering team at Google and an incoming Align MSCS candidate at Northeastern University-Seattle. He has a strong performance-oriented focus, contributing to Alibaba's BladeDISC project by enhancing TorchBench benchmarks and enabling PDLL-based model transformations to improve CPU and CUDA workloads. His background includes internships at Amazon and Google and multiple teaching assistant roles, reflecting both production experience and a commitment to mentorship. With a perfect 4.0 MSCS record and a BA in Financial Economics, he blends rigorous academic training with practical engineering, and he’s particularly fascinated by social media and internet technologies beyond his ML work.
5 years of coding experience
MS in Computer Science, 4.00/4.00, MS in Computer Science, 4.00/4.00 at Northeastern UniversityBA of Economics, Finance (Financial Economics), BA of Economics, Finance (Financial Economics) at Capital University of Economics and Business
English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (12)
benchmarking10
benchmark10
python10
compiler-compiler9
compiler9
pytorch9
deep-learning8
machine-learning8
deeplearning-ai8
mlr8
tensorflow5
cicd4
Programming languages (4)
Github contributions (5)
alibaba/BladeDISC
Nov 2022 - Jan 2023
BladeDISC is an end-to-end DynamIc Shape Compiler project for machine learning workloads.
Role in this project:
ML Engineer
ML EngineerContributions:25 reviews, 18 commits, 61 PRs in 2 months
Contributions summary:Peilin primarily contributes to the TorchBench benchmark scripts, modifying and extending test configurations for CPU and CUDA environments, demonstrating a focus on performance analysis and optimization. They are involved in setting up benchmark environments, including installing dependencies and configuring virtual environments. Additionally, the user made changes to support PDLL (Pattern Definition Language) for more general model transformations, including QGEMM and QCONV and fixing bugs in fusion patterns and folding quantize operations.
pytorchend-to-endworkloadsdeep-learningmlir
Guo-Peilin/triton
Mar 2024 - Jun 2024
Development repository for the Triton language and compiler
Contributions:13 pushes, 3 branches in 3 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.