Chunyang Wen
Software Engineer at Apple
Haidian District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolChunyang Wen is a software engineer with 12 years' experience specializing in distributed systems and AI/ML infrastructure, currently building ML platform capabilities at Apple after prior senior roles at Ant Group and Baidu. He has deep practical experience with TensorFlow and Ray-based distributed training, Kubernetes-native frameworks like ElasticDL, and production-scale optimization work in projects such as DeepSpeed and DyNet. Comfortable across C++ and Python and fluent in Linux/shell tooling, he has shipped distributed compute engines, ads-ranking pipelines, and array/ML framework improvements for Apple silicon. His contributions to high-profile open-source projects show a focus on model versioning, checkpointing, async SGD and memory/device management—areas that directly improve large-scale training reliability. Known for digging into low-level systems as well as developer-facing APIs, he pairs strong engineering rigor with a curiosity-driven approach to solve messy, real-world data and performance problems. Based in Beijing, he combines top-tier academic training with a track record of cross-company impact on ML infra and distributed computing.
12 years of coding experience
8 years of employment as a software developer
Master's degree, Computer Systems Networking and Cellular communications, Top 10% 3.76/4.0, Master's degree, Computer Systems Networking and Cellular communications, Top 10% 3.76/4.0 at University of Science and Technology of ChinaBachelor's degree, Electrical, Electronics and Communications Engineering, Bachelor's degree, Electrical, Electronics and Communications Engineering at Tianjin University
English, Chinese
Stackoverflow
244reputation
19kreached
14answers
2questions
Github Skills (35)
kubernetes10
pytorch10
c-language10
python10
gpu-programming10
machine-learning10
inference10
deepspeed10
deep-learning10
tensorflow10
kubernetes-pods10
cprogramming-language10
model-checking9
checkpoint9
refactor9
Programming languages (14)
Github contributions (5)
sql-machine-learning/elasticdl
Aug 2019 - Nov 2019
Kubernetes-native Deep Learning Framework
Role in this project:
ML Engineer
ML EngineerContributions:16 commits, 20 PRs, 8 pushes in 3 months
Contributions summary:Chunyang primarily contributed to the core functionalities of the ElasticDL framework, which is designed for deep learning on Kubernetes. Their commits focused on enhancing the master service, specifically related to model versioning, checkpointing, and gradient reporting, key components for distributed deep learning training. They also added features for asynchronous SGD and optimized the task dispatching mechanism. Furthermore, the user improved the worker's capabilities by integrating the model with the reporting infrastructure.
deep-learningmachine-learningmlopsdistributed-systemskubernetes
deepspeedai/DeepSpeed
May 2020 - Nov 2021
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Role in this project:ML Engineer
ML EngineerContributions:6 reviews, 11 commits, 13 PRs in 1 year 6 months
Contributions summary:Chunyang contributed to the DeepSpeed library, focusing on improvements to the loss scaler, refactoring, and code style unification. They addressed typos, removed redundant code, and added a logging utility. Furthermore, the user made changes to the inference engine, optimizing it for improved performance.
billion-parametersfinetuningtrainingmixture-of-expertszero
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.