Vikash Gupta
Senior Software Development Engineer at AMD
Hyderabad, Telangana, India
Summary
🤩
Rockstar🎓
Top SchoolVikash Gupta is a Senior Software Development Engineer specializing in deep learning compilers and low-power hardware accelerators, currently working at AMD after three years building ML compiler stacks at Ceremorphic. He works extensively with LLVM and MLIR, contributing backend optimizations and codegen improvements—evidenced by meaningful commits to the high-profile llvm/llvm-project around AMDGPU register allocation and Clang issues. With a BTech from IIT Bhubaneswar and a strong academic record, he combines systems-level compiler expertise with practical accelerator-aware optimization. Despite just a few years in industry, he has taken on impactful backend responsibilities that bridge research-grade compiler concepts and production hardware targets.
2 years of coding experience
4 years of employment as a software developer
Science Stream, 96% in class 12th, Science Stream, 96% in class 12th at Nirmala Convent High School, Renukut, Uttar Pradesh, IndiaBachelor of Technology - BTech, Electrical, Electronics and Communications Engineering, current CGPA : 9.07, Bachelor of Technology - BTech, Electrical, Electronics and Communications Engineering, current CGPA : 9.07 at IIT Bhubaneswar
Github Skills (12)
compiler-optimization10
code-generation10
amdgpu10
clangd10
allocation10
registers10
clang10
llvm10
cprogramming-language9
opencl9
c-language9
computer-architecture8
Programming languages (1)
Github contributions (2)
llvm/llvm-project
Apr 2024 - Apr 2025
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:80 reviews, 39 PRs, 13 pushes in 11 months
Contributions summary:Vikash primarily focused on optimizing and maintaining the LLVM compiler project. Their contributions included reverting and reapplying specific code changes related to AMDGPU (Advanced Micro Devices Graphics Processing Unit) optimization, specifically concerning SGPR spills and register allocation. They also addressed issues in the Clang compiler, specifically with __builtin_alloca and address space for OpenCL, ensuring memory allocation on the stack used the correct private address space. Additionally, they added and addressed code generation and combine pattern for fmul and select operations for AMDGPU targets.
compilerstechnologiesclangsubmittoolchain
vg0204/llvm-project
Apr 2024 - Mar 2025
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Contributions:157 pushes, 37 branches in 11 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.