Allen Li
Principal Software Engineer at NVIDIA
Tucson, Arizona, United States
Summary
🤩
Rockstar🎓
Top SchoolAllen Li is a Principal Software Engineer based in Tucson with a decade-plus career focused on GPU-accelerated computing and high-performance simulation modernization. He specializes in CUDA, OpenGL, and other GPU environments, translating complex numerical and finite-element physics codes to run efficiently on modern heterogeneous platforms. At NVIDIA he combines deep systems-level coding with product-minded feature design, and has contributed backend improvements to prominent open-source projects like NVIDIA/MatX, enhancing templated type handling, memory access, and segmented sorting support. Previously a Senior Staff Engineer at Google, he brings cross-company experience scaling performance-critical systems and shipping robust, production-ready software. He holds a master's in custom IT systems from Cal Poly San Luis Obispo and is comfortable moving between hands-on implementation and architecting platform-level solutions. A not-obvious strength: he pairs low-level algorithmic optimization with a keen sense for user-facing business impact, ensuring technical changes drive measurable customer adoption.
3 years of coding experience
9 years of employment as a software developer
California Polytechnic State University, San Luis Obispo
Github Skills (10)
cuda10
c-language10
gpgpu10
cprogramming-language10
gpu10
accelerated-computing10
cpp9
c179
pybind117
python5
Programming languages (1)
Github contributions (5)
NVIDIA/MatX
Jul 2022 - Jan 2023
An efficient C++17 GPU numerical computing library with Python-like syntax
Role in this project:
Back-end Developer
Back-end DeveloperContributions:64 reviews, 29 commits, 43 PRs in 6 months
Contributions summary:Allen primarily focused on improving the `matx` library by addressing implementation details and resolving bugs. Their commits include piping templated types through for set operations, removing unused pybind calls, and adding support for CubSortSegmented. Additionally, they refactored the code by removing deprecated shape classes and adding metadata to print, as well as addressing complex number operations. Finally, they added error handling for file reading and optimized memory access.
cudagpu-programmingnumerical-computingpythoncpp
tylera-nvidia/gpuStarterResources
Aug 2024 - Nov 2024
Contributions:13 pushes in 3 months
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.