Jorn Tuyls
Senior Member Of Technical Staff - AI Compilers at AMD
Antwerp, Antwerp, Belgium
Summary
🤩
Rockstar🎓
Top SchoolJorn Tuyls is a Senior Member of Technical Staff specializing in AI compilers with a decade of experience building and optimizing machine learning toolchains. Currently at AMD after Xilinx's acquisition, he focuses on compiler backends and GPU code generation for targets like Vulkan, Metal, ROCm and CUDA, bridging MLIR/LLVM and runtime concerns. He is an active open-source contributor to prominent projects such as Apache TVM and IREE, where his work spans TensorFlow frontend fixes, ROCm Docker integrations, Vitis-AI updates, and GPU encoding/serialization improvements. Jorn pairs production-grade engineering with research-backed knowledge from KU Leuven and work at the University of Freiburg, enabling pragmatic solutions to complex compiler and integration issues. He also brings entrepreneurial and field experience—from co-founding a student career platform to prototyping sensors for Peruvian coffee farmers—reflecting a practical, user-centered approach. Colleagues rely on him for deep technical ownership across the full ML compiler stack and for smoothing brittle integrations between frameworks and hardware.
10 years of coding experience
9 years of employment as a software developer
Master of Science Computer Science Engineering Artificial Intelligence, Master of Science Computer Science Engineering Artificial Intelligence at KU LeuvenComputer Science Research Group Machine Learning for Automated Algorithm Design, Computer Science Research Group Machine Learning for Automated Algorithm Design at The University of FreiburgASO Latijn-Wiskunde, ASO Latijn-Wiskunde at Sint-Jozefcollege Turnhout (SJT)Spanish Language, Spanish Language at CLT Leuven
English, French, Spanish, Dutch
Github Skills (19)
python10
llvm10
machine-learning10
mlr10
deeplearning-ai10
roc10
compiler-compiler10
deep-learning10
tensorflow10
compiler10
dockers9
vulkan9
docker9
gpu9
metal9
Programming languages (6)
Github contributions (5)
iree-org/iree
Apr 2024 - Apr 2026
A retargetable MLIR-based machine learning compiler and runtime toolkit.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:257 reviews, 102 PRs, 103 pushes in 2 years
Contributions summary:Jorn's contributions primarily revolve around integrating and adapting LLVM within the IREE project, specifically focusing on the MLIR component. They addressed API breaking changes, dependency issues in the Torch/InputConversion plugin, and resolved build failures. Furthermore, the user implemented and refactored GPU target attribute key strings and encoding layout interfaces, showcasing their work on optimizing code generation for different GPU targets like Vulkan, Metal, ROCM, and CUDA. The user also made changes to serialization for the `GPUEncodingLayoutAttr`.
compilermachine-learningmlirvulkantensorflow
apache/tvm
Oct 2020 - Apr 2022
Open Machine Learning Compiler Framework
Role in this project:ML Engineer & DevOps Engineer
ML Engineer & DevOps EngineerContributions:48 reviews, 12 commits, 13 PRs in 1 year 5 months
Contributions summary:Jorn's commits primarily focus on enhancing the TensorFlow frontend within the TVM compiler, specifically addressing parsing issues related to `conv2d_transpose` in TF 1.15 and adding tests to validate functionality. Additionally, the user integrated ROCm support by adding a Docker configuration for it. Furthermore, contributions include updates to the Vitis-AI integration, including fixes, documentation updates, and CI updates related to PyXIR and Vitis AI versions.
compilermachine-learningtensordeep-learninggpu
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.