Wuwei Lin
Seattle, Washington, United States
Summary
🤩
RockstarWuwei Lin is a machine learning systems engineer with 10 years of experience building and optimizing ML compilers and LLM deployment engines, currently a Member of Technical Staff at OpenAI. He has deep open-source roots as an Apache TVM committer and PMC member, contributing performance-critical backend work for CPU, GPU and accelerator targets. His recent roles at OctoAI (now NVIDIA) and NVIDIA focused on ML compiler and LLM engine co-design, quantization, and faster transformer kernels—bridging research prototypes to production-grade inference. Contributions to projects like MLC-LLM and TVM show a knack for low-level numeric optimizations (e.g., int8 GEMM and type refinements) and careful test/doc hygiene. Trained at Carnegie Mellon, he combines academic rigor with hands-on systems engineering across cloud, compiler, and model deployment stacks. Colleagues know him for quietly turning subtle type and memory changes into measurable throughput gains.
11 years of coding experience
Github Skills (25)
algorithm10
tvm10
algorithms10
c-language10
python10
machine-learning10
machine-learning-algorithms10
data-structure10
llm10
compiler-compiler10
optmization10
performance-optimization10
optimisation10
cuda10
data-structures10
Programming languages (10)
Github contributions (5)
apache/tvm
Aug 2018 - Dec 2022
Open Machine Learning Compiler Framework
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:742 reviews, 200 commits, 678 PRs in 4 years 4 months
Contributions summary:Wuwei's commits show contributions focused on optimizing the TVM deep learning compiler stack, specifically for CPU, GPU, and specialized accelerators. They implemented performance optimizations by modifying the code to use integers instead of char types and added recipes for int8 GEMM. The user also fixed bugs and updated documentation and test cases for existing convolution models. These changes indicate a strong involvement in both the backend compilation process and integration of machine learning models.
compilermachine-learningtensordeep-learninggpu
mlc-ai/mlc-llm
May 2023 - Oct 2024
Universal LLM Deployment Engine with ML Compilation
Role in this project:ML Engineer
ML EngineerContributions:30 reviews, 94 PRs, 25 pushes in 1 year 5 months
Contributions summary:Wuwei primarily contributed to the implementation and optimization of the MLC-LLM deployment engine. Their work involved fixing type hints, adding new quantization modes, and addressing issues related to the faster transformer kernel. They also worked on the model definition, specifically using SizeVar instead of Var for model definitions. Furthermore, they debugged and fixed issues regarding logit processors and cuda graphs.
llmmachine-learningmachine-learning-compilationlanguage-modeltvm
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.