Umar Arshad
AI Algorithm Engineer at Intel Corporation
Atlanta, Georgia, United States
Summary
🤩
Rockstar🎓
Top SchoolUmar Arshad is an AI Algorithm Engineer based in Atlanta with 11 years of software engineering experience, currently applying AI expertise at Intel after a decade building high-performance libraries at ArrayFire. He brings deep C++ and back-end skills, with a strong focus on correctness and test automation demonstrated by substantial validation and coverage improvements to KhronosGroup's SPIRV-Tools and bug- and memory-leak fixes in the ArrayFire GPU image library. Umar pairs academic grounding (MS and BS in Computer Science from Georgia State University) with practical teaching and industry experience, having lectured on Java and implemented production payment systems earlier in his career. Colleagues rely on him for meticulous refactors and edge-case hardening that improve long-term maintainability in performance-sensitive codebases.
11 years of coding experience
12 years of employment as a software developer
Masters, Computer Science, Masters, Computer Science at Georgia State University
Stackoverflow
980reputation
29kreached
17answers
1question
Github Skills (22)
c-language10
apidoc10
image-processing10
memory-management10
api10
cprogramming-language10
test-automation10
control-flow-graph9
opencl9
control-flow9
algorithm7
algorithms7
data-structure7
data-structures7
cuda6
Programming languages (10)
Github contributions (5)
arrayfire/arrayfire
Jul 2014 - Jan 2023
ArrayFire: a general purpose GPU library.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:7 releases, 226 reviews, 1314 commits in 8 years 7 months
Contributions summary:Umar primarily focused on updating and fixing bugs in the ArrayFire image processing library. They addressed enum naming conventions and fixed memory leaks in multiple areas. Their work included modifying and testing C++ code, updating function calls, and correcting potential edge cases within the library's API and functionalities. The user demonstrated a strong understanding of the internal workings of the image processing library as well as a concern for memory management.
cudaarrayfirecppgpuscientific-computing
KhronosGroup/SPIRV-Tools
Nov 2015 - Sep 2016
Role in this project:Back-end Developer & Test Automation Engineer
Back-end Developer & Test Automation EngineerContributions:30 commits, 36 PRs, 232 comments in 10 months
Contributions summary:Umar's contributions centered around enhancing the SPIRV-Tools validation framework. They implemented and refined logical layout validation, ensuring module-level instructions adhered to specification requirements. Additionally, they addressed identified bugs in the validation process and refactored code for improved readability and organization. A significant portion of the work focused on adding tests to cover various logical layout aspects, resulting in a substantial increase in test coverage.
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.