Wai Tsang - Principal Engineer at Intel

Wai Tsang

Principal Engineer at Intel

Markham, Ontario, Canada

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Wai Tsang is a Principal Engineer based in Markham, Ontario with seven years of focused experience in compiler development and backend engineering. He has progressed through technical roles at IBM and Intel, contributing to production compilers and rising to senior and principal levels where he drives compiler infrastructure and backend design. Wai is an active open-source contributor to notable projects like Triton and ONNX-MLIR, where his work improved LLVM translation passes, metadata provenance, and builder-based loop generation to enhance modularity and maintainability. He combines low-level compiler expertise with practical engineering practices that make complex toolchains easier to extend for new hardware backends. Collected experience translating intermediate representations and improving build metadata shows a pragmatic attention to reproducibility and long-term maintainability. A Waterloo computer science alumnus, he balances deep technical craftsmanship with a track record of shipping robust compiler features in both industry and open-source settings.

6 years of coding experience

10 years of employment as a software developer

Bachelor of Mathematics, Computer Science, Bachelor of Mathematics, Computer Science at University of Waterloo

English, Chinese, Chinese

Github Skills (12)

c-language10

onnx10

cprogramming-language10

compiler-construction10

backend10

back-end-development10

mlr10

python10

compiler-development10

llvm10

roc8

cuda8

Programming languages (6)

C++LLVMHTMLLogosMLIRPython

Github contributions (5)

onnx/onnx-mlir

Feb 2022 - Jan 2023

Representation and Reference Lowering of ONNX Models in MLIR Compiler Infrastructure

Role in this project:

Back-end Developer

Contributions:64 reviews, 32 commits, 51 PRs in 11 months

Contributions summary:Wai primarily contributed to the core functionality of the ONNX-MLIR project, with a focus on improving the compiler's capabilities. They implemented features to emit version metadata and include Git commit IDs within the LLVM metadata, ensuring the project's build information is readily available. Additionally, the user refactored code to utilize builder-based interfaces for generating Krnl loops, improving the code's maintainability and potentially its performance. The user's commits consistently improved the efficiency and reliability of the ONNX-MLIR compiler.

pytorchrepresentationdeep-learningmlironnx-models

triton-lang/triton

Aug 2023 - Jan 2025

Development repository for the Triton language and compiler

Role in this project:

Backend Developer

Contributions:15 reviews, 16 PRs, 8 pushes in 1 year 5 months

Contributions summary:Wai primarily contributed to the Triton compiler, focusing on improvements and modifications to the TritonGPUToLLVM pass. Their work involved correcting option passing, modifying target selection via an enum, and generalizing tests for various backends. The user also made changes to the LLVMIRTranslation and python interfaces, further demonstrating involvement in the backend infrastructure and the compilation process. These changes were focused on improving the modularity, and maintainability of the compilation process, and making it more flexible to support new backends.

compilerprogramming-languagecode-generationtriton

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.

Request Free Trial