Kai Yang
Staff Engineer 高级技术专家 at Ant Group
Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolKai Yang is a Staff Engineer with 14 years of experience designing and optimizing distributed systems and ML infrastructure, currently driving core Ray development at Ant Group. He has deep cross-language systems expertise—C++, Java, and Python—and has led performance, reliability, and refactoring efforts across RPC, object stores, and worker implementations. A long-standing Ray committer, he contributed fixes and architectural changes to the popular ray-project/ray engine, including worker limits, GCS data-structure improvements, and Redis-related robustness fixes. His background includes building production-grade distributed platforms at Microsoft and startups, from real-time agent systems to big-data workbenches integrating Spark, Kubernetes, and Airflow. Known for pragmatic engineering excellence, he pairs low-level optimization (zero-copy buffers, reference counting) with deployment and multi-tenant operational improvements. Based in Beijing, he blends deep systems craftsmanship with a pragmatic focus on reliability and large-scale ML workloads.
14 years of coding experience
4 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at University of Science and Technology of China
Github Skills (11)
distributed-systems10
c-language10
ray10
cprogramming-language10
redis10
concurrency9
data-structure9
management9
data-structures9
manage9
python7
Programming languages (14)
Github contributions (5)
ray-project/ray
Jan 2019 - Jul 2022
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:703 reviews, 140 commits, 320 PRs in 3 years 6 months
Contributions summary:Kai's contributions focused on improving the Ray AI compute engine's core functionalities. They implemented limits on the maximum number of workers started per language to optimize resource usage. The user also introduced and refactored the data structures within the GCS, changing object tables to Set instances to better manage data integrity. Furthermore, the user addressed several bugs related to Redis interaction in relation to object management and implemented fixes for potential worker process failures.
pythonconsistsruntimetensorflowserving
kfstorm/DoubanFM
Jul 2011 - Jun 2014
Contributions:1 release, 308 commits, 2 PRs in 2 years 10 months
douban-fmwindowsdouban
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.