Jun Yang
Senior Director Of Compute Arch at NVIDIA
Chaoyang District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolJun Yang is a seasoned compute architect and engineering leader with 13 years of experience designing and optimizing AI systems across software and hardware stacks, currently serving as Senior Director of Compute Architecture at NVIDIA. He leads end-to-end AI optimization—spanning compilers, distributed systems, GPU kernel tuning, model compression and cluster resource utilization—often applying system-and-model co-design to squeeze more performance with less compute. Prior roles at Alibaba Cloud, Qihoo 360, Yahoo and Cadence show deep expertise in production-scale inference/training pipelines, recommendation systems, and compiler/frontend development. Jun is a hands-on contributor to core open-source projects like Redis/KeyDB, where he improved stream command robustness and CLI security—evidence of his focus on reliability and usability in low-level systems. Based in Beijing, he combines academic grounding in computer architecture with a track record of beating vendor implementations (e.g., cuDNN/cuBLAS/TensorRT) through careful engineering and algorithmic tuning.
13 years of coding experience
16 years of employment as a software developer
Bachelor's degree Computer Science, Bachelor's degree Computer Science at The University of Science and Technology of ChinaMaster's degree Computer Architecture, Master's degree Computer Architecture at Chinese Academy of Sciences
Github Skills (19)
key-value-store10
redis10
command-line-interface10
databases10
c1110
c1710
command-line10
stream-processing10
database10
cli10
key-value9
error-handling9
nosql8
message-broker7
object-cache6
Programming languages (5)
Github contributions (5)
Snapchat/KeyDB
Apr 2018 - Aug 2018
A Multithreaded Fork of Redis
Role in this project:
Back-end Developer
Back-end DeveloperContributions:60 commits in 4 months
Contributions summary:Jun primarily contributed to bug fixes within the KeyDB codebase, focusing on the XADD command, and redis-cli utilities. The fixes addressed issues related to argument handling, parsing of IDs, and the command-line interface. The user's contributions improved the robustness of the stream processing and the security of the command-line tool by adding warnings.
redis
redis/redis
Apr 2018 - Aug 2018
For developers, who are building real-time data-driven applications, Redis is the preferred, fastest, and most feature-rich cache, data structure server, and document and vector query engine.
Role in this project:Back-end Developer
Back-end DeveloperContributions:42 commits, 27 PRs, 63 comments in 4 months
Contributions summary:Jun primarily focused on bug fixes and improvements within the Redis codebase, specifically targeting the XADD command and redis-cli tool. Their contributions involved correcting arity checks, addressing issues with ID parsing and repeat command logic, and fixing memory leaks. Furthermore, the user added warning messages, refined existing messages, and updated the command-line interface for security and usability. These changes reflect a focus on stability, security, and user experience improvements to the Redis CLI.
data-structurequery-engineredisdatabasekey-value
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.