Chao Mei
Software Engineer at Google
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolChao Mei is a software engineer based in Mountain View with seven years of experience building high-performance systems at Google and prior work at Intel exploring exascale programming models. He focuses on efficient ML inference, contributing to the widely used google/XNNPACK project where he implemented memory-reduction strategies like a "Greedy by size planner" and improved code robustness for mobile and server neural network workloads. His background in computer science from Fudan and a Ph.D. program at UIUC underpin a research-informed approach to pragmatic engineering problems. Colleagues describe him as detail-oriented and quality-driven, demonstrated by careful fixes, macro guards, and benchmark adjustments that keep production tooling stable.
7 years of coding experience
1 year of employment as a software developer
B.S., Computer Science, B.S., Computer Science at Fudan UniversityUniversity of Illinois Urbana-Champaign
Github Skills (8)
neural-network10
c1710
memory-optimization10
simd10
inference10
c1110
multithreading9
matrix-multiplication8
Programming languages (4)
Github contributions (5)
google/XNNPACK
Nov 2019 - May 2021
High-efficiency floating-point neural network inference operators for mobile, server, and Web
Role in this project:
ML Engineer
ML EngineerContributions:6 commits in 1 year 6 months
Contributions summary:Chao's commits focus on optimizing the XNNPACK library for high-efficiency neural network inference. They implemented memory optimization techniques, specifically a "Greedy by size planner," to reduce the memory footprint of intermediate tensors. Further contributions include fixing typos and guarding code segments with macros to prevent compilation errors, indicating a focus on code quality and maintainability within the context of the inference library. The user also disabled the application of the XNNPACK delegate for TFLite benchmarks within the XNNPACK/bench directory.
multithreadingsimdtensorflowcpufloating
multiverse-tf/gemmlowp
Jul 2019 - Jul 2019
Low-precision matrix multiplication
Contributions:1 push in 1 day
matrix-multiplicationlinear-algebraprecisionmultiplicationmatrix
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.