Ze Liu
Member Of Technical Staff at xAI
Palo Alto, California, United States
Summary
🤩
Rockstar🎓
Top SchoolZe Liu is a Member of Technical Staff at xAI with seven years of experience building cutting-edge AI systems, currently contributing to Grok Imagine, Vision, and Voice modes. He holds a PhD from the University of Science and Technology of China and was a Microsoft Research PhD Fellow and research intern for several years, blending deep academic rigor with product-focused engineering. Ze made notable open-source contributions to the widely used Swin Transformer repository, originating core model files and adding innovations like a Swin MLP and ImageNet-22K pretrained support. Based in Palo Alto, he brings a rare combination of top-ranked academic performance and practical ML engineering that spans research prototypes to production-ready model improvements. An attentive implementer, he often authors foundational commits and build tooling changes that enable higher-resolution fine-tuning and smoother validation workflows.
7 years of coding experience
Doctor of Philosophy - PhD, Doctor of Philosophy - PhD at University of Science and Technology of China
Github Skills (11)
swin-transformer10
computer-vision10
pytorch10
machine-learning10
deep-learning10
modeling10
image-classification10
implement10
data-science9
python9
object-detection9
Programming languages (1)
Github contributions (5)
microsoft/Swin-Transformer
Mar 2021 - Oct 2022
This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".
Role in this project:
ML Engineer
ML EngineerContributions:22 commits, 24 PRs, 23 pushes in 1 year 6 months
Contributions summary:Ze contributed significantly to the Swin Transformer implementation, introducing core components and functionalities. They initiated the project with an initial commit containing core model files and later added a novel "Swin MLP" architecture. Additionally, they incorporated support for fine-tuning on higher resolution data and integrated the Swin Transformer V2 model. Furthermore, the user modified build files to support different validation samplers and added pre-trained weights from the ImageNet-22K dataset and various bug fixes.
semantic-segmentationhierarchicalvisual-recognitionswinimage-classification
zeliu98/CloserLook3D
Jul 2020 - Dec 2020
A Closer Look at Local Aggregation Operators in Point Cloud Analysis(ECCV 2020)
Contributions:11 commits, 4 PRs, 9 pushes in 5 months
lookpoint-cloudsaggregationeccvpoint
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.