Yu-chen Cheng
Software Engineer at Tencent
Shanghai, Shanghai, China
Summary
🤩
Rockstar🎓
Top SchoolYu-Chen Cheng is a software engineer with nine years' experience specializing in AI/LLM infrastructure and Kubernetes-based MLOps. Currently at Tencent Cloud TI-ONE, he builds scalable inference and training platforms informed by prior roles developing internal ML training systems and LLMOps at startups and enterprises. He holds an M.Eng. and B.Eng. in Software Engineering from Shanghai Jiao Tong University and has tutored multiple software courses there, showing a commitment to engineering education. As an open-source mentee on the CNCF volcano project, he improved e2e testing stability and modernized test tooling—work that underscores a focus on reliability and reproducible CI for cloud-native batch workloads. Comfortable across SRE, DevOps, and platform engineering, he blends hands-on testing and automation improvements with production-facing architecture. Based in Shanghai, he combines academic rigor with practical experience shipping AI infrastructure at scale.
9 years of coding experience
2 years of employment as a software developer
Master of Engineering - MEng Software Engineering, Master of Engineering - MEng Software Engineering at Shanghai Jiao Tong University
Stackoverflow
46reputation
5kreached
4answers
2questions
Github Skills (20)
kubernetes10
testing10
cicd10
e2e10
kubernetes-pods10
batchfile9
go9
batch-processing9
batch9
docker8
dockers8
bash7
kind6
microservices6
swift6
Programming languages (13)
Github contributions (5)
volcano-sh/volcano
Feb 2021 - Aug 2021
A Cloud Native Batch System (Project under CNCF)
Role in this project:
DevOps Engineer
DevOps EngineerContributions:2 reviews, 8 commits, 5 PRs in 5 months
Contributions summary:Yu-chen's contributions primarily focus on improving the e2e (end-to-end) testing infrastructure and scripts for the volcano project. They addressed compatibility issues by replacing deprecated commands related to `kind` (Kubernetes in Docker) within the e2e test scripts. Furthermore, they refactored e2e tests to improve stability, readability and imported a stress test. These changes indicate a focus on enhancing the reliability and maintainability of the project's testing framework.
golangbigdatabatchmachine-learningkubernetes
dyweb/go-jaccount
Mar 2020 - Jun 2022
A Go Client for jAccount
Contributions:1 release, 2 reviews, 22 commits in 2 years 3 months
golanggo-clientgojaccountjava
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.