Yuan Zhou
Engineering Manager at IBM
Dublin, Ireland
Summary
🤩
Rockstar🎓
Top SchoolYuan Zhou is an engineering manager with 13 years of experience specializing in performance engineering for databases, virtualization, cloud storage, and Big Data, now leading Spark and Gluten integration at IBM Watsonx. He combines deep systems-level expertise in C/C++, Scala, and Python with practical DevOps and CI/CD skills, having improved pipelines and performance for high-profile open-source projects like OpenStack Swift and Apache Arrow. Yuan has recently pivoted into LLM infrastructure and applications—optimizing vLLM CPU backends and building Text-to-SQL and UDF codegen solutions—bringing uncommon cross-domain fluency between Big Data kernels and generative AI stacks. He has a track record of shipping kernel-level optimizations and pragmatic tooling (e.g., mmap file I/O, HDFS Kerberos fixes, DB auditor tuning) that measurably improve throughput and reliability. Based in Dublin, he blends research-rooted rigor from Tongji University with hands-on engineering leadership across Intel and IBM, often focusing on production-grade performance that accelerates customer adoption.
13 years of coding experience
14 years of employment as a software developer
Master Computer Science, Master Computer Science at Tongji University
Chinese, English
Github Skills (46)
github-ci10
docker10
c-language10
back-end-development10
python10
bash10
hdfs10
object-storage10
dockers10
cicd10
apache-arrow10
bash-shell10
script10
openstack10
optimisation10
Programming languages (8)
Github contributions (5)
apache/incubator-gluten
Apr 2022 - Jan 2023
Gluten is a middle layer responsible for offloading JVM-based SQL engines' execution to native engines.
Role in this project:
DevOps Engineer
DevOps EngineerContributions:808 reviews, 44 commits, 800 PRs in 9 months
Contributions summary:Yuan primarily focused on improving the repository's CI/CD pipeline. Their contributions include adding and configuring GitHub Actions to automate tasks like code formatting, unit tests, and performance tests triggered by pull requests. They also addressed build issues and dependency installations, including libhdfs3 and S3 support. Furthermore, they refined scripts related to build processes, demonstrating a focus on infrastructure and build automation.
arrowclickhousesimdspark-sqlvectorization
vllm-project/vllm
Jun 2023 - Mar 2025
A high-throughput and memory-efficient inference and serving engine for LLMs
Role in this project:MLOps Engineer
MLOps EngineerContributions:31 reviews, 14 PRs, 97 comments in 1 year 9 months
Contributions summary:Yuan primarily contributed to improving the CI/CD pipeline and enhancing CPU backend performance within the vLLM project. They added Intel OpenMP tunings and incorporated a proxy for HTTP connections in the Dockerfile, alongside refining the CPU test infrastructure to allow binding to different cores and incorporating the build number in image names. Furthermore, the user updated documentation to reflect CPU backend considerations and usage, and refactored CPU tests for better core binding. These contributions directly improve the efficiency, deployment, and testing of the LLM serving engine on CPU platforms.
amdcudadeepseekgpthpu
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.