Mike Wilson
Senior Systems Software Engineer at NVIDIA
Asheville, North Carolina, United States
Summary
🤩
Rockstar🎓
Top SchoolMike Wilson is a Senior Systems Software Engineer based in Asheville, NC with 12 years of experience specializing in embedded systems and low-level platforms. He currently works at NVIDIA contributing to high-performance back-end work on cuDF, refactoring and optimizing GPU DataFrame internals and improving serialization and ORC write consistency. His background spans microcontrollers (AVR, ARM), console/SPU work on PS3 and Xbox 360, game-engine streaming and filesystem tooling, and earlier Linux device-driver and kernel bring-up at Motorola. That unusual blend of embedded/console systems, GPU data-performance engineering, and Kubernetes/cloud deployment experience gives him a rare ability to bridge hardware-focused optimization and scalable software infrastructure.
12 years of coding experience
17 years of employment as a software developer
Appalachian State University
Github Skills (11)
parquet10
cuda10
c-language10
performance-optimization10
c-programming-language10
orc9
data-structure9
strings9
text-manipulation9
byte-array9
data-structures9
Programming languages (23)
Github contributions (5)
rapidsai/cudf
Aug 2020 - Jan 2023
cuDF - GPU DataFrame Library
Role in this project:
Back-end Developer
Back-end DeveloperContributions:899 reviews, 82 commits, 71 PRs in 2 years 5 months
Contributions summary:Mike primarily focused on refactoring and optimizing the cuDF codebase, specifically addressing performance issues and code deprecation warnings. Their work involved replacing deprecated functions with their recommended alternatives, updating the codebase to align with current coding standards, and implementing a new mechanism for converting data between row-major and column-major formats. The changes indicate a focus on maintaining and improving the efficiency of the library's core functionalities. Additionally, the user implemented features to add string and byte array columns into the orc files, while ensuring that statistics were accurately persisted across multiple write calls.
cudadataframe-librarydata-analysiscppcudf
hyperbolic2346/rmm
Sep 2020 - Jul 2023
RAPIDS Memory Manager
Contributions:40 pushes, 5 branches in 2 years 10 months
memorydeltamemory-allocationmemory-managerrapids
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.