Ivan Kosarev is a compiler engineer with nine years of focused experience building and refining development toolchains for high-performance and heterogeneous architectures. Currently at AMD and formerly the LLVM lead for Imagination Technologies, he has delivered end-to-end compiler features from front-end language extensions to low-level code generation and optimizations. A long-time LLVM contributor, he improved the AMDGPU assembly parser and upstreamed type-based alias analysis work that unlocked scalar optimizations across projects. He’s consulted for major vendors designing scalable, maintenance-friendly LLVM-based stacks and has practical experience porting sanitizers and integrating toolchains with OS kernels like FreeBSD. Based in Leighton Buzzard, he combines deep compiler internals expertise with a track record of pragmatic engineering and collaboration on widely used open-source projects.
9 years of coding experience
1 year of employment as a software developer
Master of Science - MS, High Performance Computing, Master of Science - MS, High Performance Computing at The University of Edinburgh
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:
Back-end Developer
Contributions:239 reviews, 133 PRs, 175 pushes in 3 years 2 months
Contributions summary:Ivan primarily contributed to the AMDGPU assembly parser within the LLVM project, focusing on improving the handling and validation of operands, specifically for AMDGPU architecture. Their work included modifying the parsing logic to automatically generate predicates, eliminating unused code, and correcting errors in offset and wait parameter validation. They also addressed issues with section type handling in the MCAsmStreamer and added support for UC_VERSION constants within the disassembler.
ZX Spectrum emulator written in a mix of Python and C++
Contributions:438 commits, 530 pushes, 1 branch in 4 years 1 month
emulatorpythonzx-spectrum-emulatorspectrummix
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.