Garry Fang
Hangzhou City, Zhejiang, China
Summary
🤩
Rockstar🎓
Top SchoolGarry Fang is an AI infrastructure engineer with nine years of experience building accelerator-to-cluster systems and large-scale offline computing platforms, currently working across infrastructure at Ant Group and Tencent. He specializes in multi-cluster Kubernetes orchestration and scheduling, contributing as a committer and reviewer to prominent CNCF projects like Karmada and Clusternet and to scheduler ecosystems including Crane and Scheduler-Plugins. His work blends backend and DevOps expertise—improving cluster lifecycle, replica scheduling, taints/tolerations, and topology-aware scheduling—while also tackling GPU training and AIOps use cases. A Zhejiang University MEng and top-ranked Sichuan University alumnus, he is an active open-source speaker and lecturer who brings production-hardened code and system-level design to complex cloud-native problems.
9 years of coding experience
Bachelor of Engineering - BE, Electrical Engineering, Top 1%, Bachelor of Engineering - BE, Electrical Engineering, Top 1% at Sichuan UniversityMaster of Engineering - MEng, Electrical Engineering, Master of Engineering - MEng, Electrical Engineering at Zhejiang University
Github Skills (20)
algorithms10
kubernetes10
apidoc10
resource-management10
cicd10
scheduler10
scheduling10
go10
api10
sched10
kubernetes-pods10
cluster-manager10
cluster-api10
orchestration10
orchestra10
Programming languages (10)
Github contributions (5)
karmada-io/karmada
May 2021 - Dec 2022
Open, Multi-Cloud, Multi-Cluster Kubernetes Orchestration
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:481 reviews, 141 commits, 105 PRs in 1 year 7 months
Contributions summary:Garry primarily contributed to the development and maintenance of the Karmada project, focusing on features related to cluster status monitoring, replica scheduling, and overall system stability. They were involved in fixing bugs related to plugin memory addresses and workload name conflicts. They also worked on enhancing cluster API, including features like ResourceSummary and also implemented and tested cluster-level lifecycle management components and incorporated the necessary code generation steps. The user also made changes to the scheduler, adding code generation, and adding the option for the scheduler estimator and improving code for a more comprehensive architecture.
containerscloud-computingmulti-clusterorchestrationcloud
clusternet/clusternet
Dec 2021 - Nov 2022
[CNCF Sandbox Project] Managing your Kubernetes clusters (including public, private, edge, etc.) as easily as visiting the Internet
Role in this project:Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:29 reviews, 26 commits, 21 PRs in 10 months
Contributions summary:Garry primarily contributed to adding features related to taints and tolerations, and the implementation of dynamic replica dividing. The user added APIs for dividing scheduling, generated necessary files, and also optimized the scheduler framework. Furthermore, the user implemented a local running script to immediately establish a local environment for the project, including the installation and configuration of Clusternet components.
orchestrationmulti-clusterclusterscluster-managementout-of-the-box
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.