Dohyun Kim
Compiler Engineer at FuriosaAI
Seoul, South Korea
Summary
🤩
Rockstar🎓
Top SchoolDohyun Kim is a compiler engineer with a decade of software experience, currently optimizing DMA code generation at FuriosaAI's NPU compiler team. He previously improved GPU runtime performance on Intel's OpenVINO project—enabling blocked data formats and optimizing primitives—bringing practical backend and performance engineering to production AI inference stacks. His background spans Rust and C++ compiler work, deep learning runtimes, and systems-level game and graphics projects, reflecting a rare mix of low-level code generation and higher-level ML engineering. A competitive programmer with ICPC World Finals honorable mention and multiple national contest wins, he pairs algorithmic rigor with production-focused optimization. Fluent in Korean and comfortable in English, he also served in a cyber operations unit during his military service, which contributed to his disciplined approach to secure, performant software.
10 years of coding experience
1 year of employment as a software developer
고등학교, 컴퓨터게임개발과, 고등학교, 컴퓨터게임개발과 at 한국게임과학고등학교Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at 숭실대학교
Github Skills (9)
c-language10
cprogramming-language10
gpu10
performance-optimization10
openvino10
deep-learning8
inference8
computer-vision8
ai8
Programming languages (3)
Github contributions (5)
openvinotoolkit/openvino
Jun 2022 - Jan 2023
OpenVINO™ is an open source toolkit for optimizing and deploying AI inference
Role in this project:
Backend & Performance Engineer
Backend & Performance EngineerContributions:164 reviews, 33 commits, 64 PRs in 7 months
Contributions summary:Dohyun primarily contributed to the OpenVINO toolkit by enabling and optimizing functionalities within the Intel GPU plugin. Their work involved enabling blocked data formats in primitives like Border and Gather, supporting non-default planar formats, and optimizing the DepthToSpace primitive. The contributions included adding unit tests for validating these changes, and refactoring existing code.
inference-enginepytorchmodel-optimizerdeep-learninggpu
wolfrev0/PER-D3QN
Mar 2021 - Aug 2022
Contributions:22 commits, 11 pushes in 1 year 5 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.