Yiran Wang
Member Of Technical Staff at Microsoft AI
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolYiran Wang is a seasoned infrastructure and compute engineer with 11 years of experience building scalable container and orchestration systems, currently focused on scaling at OpenAI after leading compute scheduling and orchestration for Microsoft AI. He’s driven production-grade infra at Character.AI, Airbnb, Anyscale and Uber, where he led open-source projects Makisu and Kraken—tools that sped Docker image builds and enabled P2P registry distribution at massive scale. Comfortable as an IC and manager, he blends low-level back-end/DevOps craftsmanship (storage layers, Dockerfiles, CI) with team leadership overseeing supercomputing inference and training stacks. His work repeatedly targets reproducible, unprivileged container workflows and hybrid cloud migrations, reflecting deep practical expertise in container image build systems and scheduler design. Based in San Francisco with a CS/math background from UIUC, he pairs academic rigor with an unusual track record of shipping both open-source tooling and internal control-plane services for ML workloads.
11 years of coding experience
15 years of employment as a software developer
Henan Experimental Middle SchoolBachelor of Science (BS) Mathematics and Computer Science, Bachelor of Science (BS) Mathematics and Computer Science at University of Illinois Urbana-Champaign
Chinese
Github Skills (18)
caching10
kubernetes10
filesystem10
docker10
bash10
containerization10
dockers10
cicd10
go10
kubernetes-pods10
back-end-development9
github-ci9
automated-build9
githubaction-workflow9
bittorrent8
Programming languages (3)
Github contributions (5)
uber-archive/makisu
Nov 2018 - Mar 2021
Fast and flexible Docker image building tool, works in unprivileged containerized environments like Mesos and Kubernetes.
Role in this project:
Back-end Developer & DevOps Engineer
Back-end Developer & DevOps EngineerContributions:6 releases, 3 reviews, 70 commits in 2 years 4 months
Contributions summary:Yiran contributed to the core functionality of the Makisu Docker image building tool by implementing changes to the underlying storage layer and file management. Their work includes updates to the file map implementation and the addition of a local key-value cache store. Further contributions encompass enhancements related to build configuration, including handling command-line options and error handling. The user also addressed Mac-specific compatibility issues and improved overall code quality.
containersdocker-imageunprivilegedcontainerizeddocker
uber/kraken
Dec 2018 - May 2019
P2P Docker registry capable of distributing TBs of data in seconds
Role in this project:DevOps Engineer
DevOps EngineerContributions:4 releases, 52 commits, 80 PRs in 5 months
Contributions summary:Yiran's primary contributions revolve around configuring and maintaining the build and deployment infrastructure for the "kraken" project. They added and modified Travis CI configurations, created scripts for starting development components using Docker, and updated scripts to start devcluster setups. These changes involved modifying Dockerfiles and configurations to ensure smooth and reproducible builds and deployments of the Kraken P2P Docker registry.
docker-imagep2pbittorrentdockertbs
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.