Zichuan Wei
Software Development Engineer at Google
Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolZichuan Wei is a Software Development Engineer based in Toronto with eight years of experience building ML and systems software, currently contributing to TensorFlow Lite at Google. He has practical ML engineering experience from Qualcomm and significant research exposure in quantum machine learning from internships and academic roles. At Google he has contributed notable open-source work to the high-profile tensorflow/tensorflow repo, adding StableHLO conversion, serialization support for operators, and improvements that enable larger model generation. He blends compiler and backend experience (IBM) with hands-on debugging and operational support from earlier IT roles, giving him a strong systems-first approach to production ML tooling. Zichuan holds a Mathematics and Computer Science degree from the University of Waterloo and brings a rare mix of research intuition and production-focused implementation skills.
7 years of coding experience
3 years of employment as a software developer
Bachelor’s Degree, Mathematics and Computer Science, 79%, Bachelor’s Degree, Mathematics and Computer Science, 79% at University of Waterloo
Chinese
Github Skills (8)
machine-learning10
tensorflow-lite10
model-conversion10
stable10
python9
deep-learning9
c-language8
cprogramming-language8
Programming languages (2)
Github contributions (5)
tensorflow/tensorflow
Aug 2022 - Dec 2022
An Open Source Machine Learning Framework for Everyone
Role in this project:
ML Engineer
ML EngineerContributions:8 reviews, 9 commits, 10 comments in 3 months
Contributions summary:Zichuan's commits primarily focus on enhancing the TensorFlow Lite (TFLite) framework to incorporate StableHLO operations. Their work includes adding experimental flags for StableHLO conversion within the TFLite conversion pipeline and integrating StableHLO serialization support for several operators like convolution, reduce_window, and custom calls, enabling models to be converted to stablehlo format. They also added support for data type conversions and improvements to the 2GB model generation.
pythondata-sciencedeep-learningmlmachine-learning
zichuan-wei/crypto_chanllenge
May 2018 - May 2019
Contributions:5 pushes, 1 branch in 11 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.