Yiyuan Zhang

Graduate Student

Hong Kong Island, Hong Kong, China
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
🎓
Top School
Yiyuan Zhang is a graduate student and Ph.D. researcher based in Hong Kong with five years of hands-on experience in ML systems and MLOps. At Zhejiang University and MMLab he focuses on making large-model training and deployment more efficient, contributing to tooling that adds FSDP, gradient checkpointing, and push-to-hub workflows. He’s practical in refining training scripts and configurations—tuning learning rates, batch sizes, and output pipelines—to move research prototypes toward reproducible, production-ready runs. Colleagues note his collaborative approach to code refinement and his knack for streamlining complex training setups.
code5 years of coding experience
bookZhejiang University
github-logo-circle

Github Skills (14)

pytorch10
mlops10
trainings10
python10
modeling10
fs9
machine-learning9
bash9
script9
qx8
qq8
git5
docker4
dockers4

Programming languages (1)

Python

Github contributions (5)

github-logo-circle
simplescaling/s1

Feb 2025 - Feb 2025

s1: Simple test-time scaling
Role in this project:
userMLOps Engineer
Contributions:5 reviews, 1 PR, 6 comments in 1 day
Contributions summary:Yiyuan primarily focused on modifying training scripts and configurations for a machine learning model. Their contributions include updating training parameters like learning rates, batch sizes, and output directories. They also worked on incorporating features such as FSDP, gradient checkpointing, and push-to-hub functionalities, indicating an effort to streamline model training and deployment processes. Additionally, the user collaborated with another developer on refining the training script and commenting on the code.
AILab-CVC/UniRepLKNet

Nov 2023 - Oct 2024

UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition
Contributions:1 PR, 6 pushes, 12 comments in 11 months
architectureartificial-intelligenceconvolutional-neural-networksdeep-learningmultimodal-learning
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.
Request Free Trial