Pierre Van Houtryve
Member Of Technical Staff at AMD
Mons, Hainaut, Belgium
Summary
🤩
Rockstar🎓
Top SchoolPierre Van Houtryve is a C++ compiler engineer with eight years of experience focused on GPU backends and low-level code generation, currently a Member of Technical Staff at AMD working on the AMDGPU backend in LLVM. He specializes in GPU compute memory models—analyzing hardware, implementing compiler codegen, and documenting models so developers can write correct and performant GPU code. His open-source contributions to the high-profile llvm-project include fixes and enhancements for the AMDGPU target, atomic optimizations, dynamic LDS linkage, and target-specific LTO improvements. Prior roles include machine-code obfuscation and optimization work at Guardsquare and ARM-focused LLVM improvements during an internship, reflecting deep familiarity with x64, ARMv8-A, and vector extensions. Colleagues rely on him to bridge hardware details and compiler implementation, turning subtle architectural behaviors into robust compiler semantics. Based in Mons, Belgium, he combines pragmatic engineering with a taste for challenging, architecture-near problems that benefit both internal teams and the wider open-source ecosystem.
8 years of coding experience
4 years of employment as a software developer
Bachelor's degree Business Computing, Bachelor's degree Business Computing at Haute Ecole Provinciale de Hainaut - Condorcet
English, French, Spanish
Stackoverflow
11reputation
3kreached
0answers
2questions
Github Skills (9)
compiler-optimization10
c-language10
amdgpu10
cprogramming-language10
llvm10
parallel-computing9
assembly8
objective-c6
macos6
Programming languages (5)
Github contributions (5)
llvm/llvm-project
Feb 2020 - Apr 2025
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:469 reviews, 253 PRs, 321 pushes in 5 years 2 months
Contributions summary:Pierre made significant contributions to the LLVM project, focusing on the AMDGPU target. Their commits involved fixing atomic optimization issues related to divergent values, specifically addressing DT updates and edge cases in the iterative strategy of the atomic optimizer. They also worked on allowing different linkage types for dynamic LDS variables, and on target-specific module splitting, enhancing the LTO functionality for AMDGPU.
compilerstechnologiesclangsubmittoolchain
Pierre-vh/Fox
Nov 2017 - Jul 2019
Contributions:3760 commits, 1 PR, 1417 pushes in 1 year 8 months
easy-to-learnin-progresscompilerlearn-programmingprogramming-language
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.