Annop Wongwathanarat
Staff Software Engineer at Arm
Manchester, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolAnnop Wongwathanarat is a Staff Software Engineer based in Manchester with a PhD in Physics and a decade-long background bridging numerical simulations, high-performance computing, and code optimization for distributed-memory systems. He has driven production-grade performance work at Arm—leading FFT optimization in Arm Performance Libraries and optimizing generative ML runtimes on AArch64—while maintaining Spack packages that integrate widely used scientific codes. His open-source contributions include performance-focused enhancements to PyTorch’s quantized operations and substantive backend work on the popular spack/spack package manager, reflecting a rare mix of low-level SIMD tuning and large-scale packaging/validation. Previously he translated academic research into robust simulation and DevOps tooling at institutes like Max Planck, RIKEN, and DTU, demonstrating an ability to move from theory to production. Colleagues value him for combining rigorous scientific training with pragmatic engineering that measurably improves compute efficiency on real hardware.
7 years of coding experience
13 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Physics, Doctor of Philosophy (Ph.D.) Physics at Technical University of Munich
English, Thai
Github Skills (18)
pytorch10
package-management10
python10
machine-learning10
hpc10
deep-learning10
gpu10
neural-network10
quantization10
build-tools10
linux9
autograd9
tensor9
arm9
numpy8
Programming languages (4)
Github contributions (5)
spack/spack
Apr 2022 - Jan 2023
A flexible package manager that supports multiple versions, configurations, platforms, and compilers.
Role in this project:
Backend Developer
Backend DeveloperContributions:30 reviews, 15 commits, 45 PRs in 8 months
Contributions summary:Annop primarily contributes to the `spack/spack` repository by adding and modifying package definitions, specifically focusing on the Arm Performance Libraries (Armpl) and Allinea Compiler (acfl) packages. Their commits include adding new versions of these libraries, updating checksums, and integrating them with other packages like Gromacs, Quantum-Espresso, CP2K, and QMCPACK. Furthermore, the user has added post-installation checks, indicating a focus on ensuring the correct functionality of the packages.
compilerspythonradiussplatformslinux
pytorch/pytorch
Sep 2024 - Mar 2025
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:ML Engineer
ML EngineerContributions:16 reviews, 7 PRs, 65 comments in 6 months
Contributions summary:Annop's contributions primarily revolve around enhancing the PyTorch framework's capabilities in the realm of quantized neural networks. They focused on optimizing and extending support for quantized operations, specifically targeting 8-bit quantization. Their work included adding NEON implementations for quantized embedding bags on aarch64 architecture and enabling s8s8s8 for qlinear with mkl-dnn, directly improving performance. They also corrected errors in existing code related to quantization, ensuring accuracy.
pythongpu-accelerationdeep-learninggpunumpy
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.