Zuobai Zhang
Montreal, Quebec, Canada
Summary
🤩
Rockstar🎓
Top SchoolZuobai Zhang is a research scientist and fourth-year PhD candidate at Mila with seven years of experience building machine learning systems for drug discovery and recommender products. Now at NVIDIA Research, he focuses on deep generative learning while previously contributing to production recommender algorithms at ByteDance. His open-source work on torchdrug reflects practical expertise in graph-based molecular models and protein dataset integration, including fixes and model improvements for SchNet and dataset loaders like ZINC2m and ChEMBL. Comfortable bridging research and engineering, he brings both academic rigor and hands-on debugging experience that improves core ML pipelines.
7 years of coding experience
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Mila - Quebec Artificial Intelligence InstituteBachelor's degree, Computer Science, Bachelor's degree, Computer Science at Fudan University
Github Skills (11)
pytorch10
drug-discovery10
deep-learning10
graph-neural-network10
python10
machine-learning9
datastructures-algorithms8
algorithms8
data-structures8
algorithm8
data-structure8
Programming languages (4)
Github contributions (5)
DeepGraphLearning/torchdrug
Aug 2021 - Nov 2022
A powerful and flexible machine learning platform for drug discovery
Role in this project:
ML Engineer
ML EngineerContributions:19 commits, 3 PRs, 16 pushes in 1 year 3 months
Contributions summary:Zuobai primarily contributed to bug fixes and improvements within the `torchdrug` repository, focusing on the core deep learning functionalities. Their work involved addressing errors related to data handling, specifically in `dataset.py` and the loading of specific datasets like ZINC2m and ChEMBL. They also made improvements to existing models (SchNet) including adding test code, as well as adding support and classes for protein datasets, indicating a focus on model development and model integration with other modules.
pytorchmachine-learning-platformlearning-platformdataminingdrug-discovery
DeepGraphLearning/GearNet
Jul 2022 - Feb 2023
GearNet and Geometric Pretraining Methods for Protein Structure Representation Learning, ICLR'2023 (https://arxiv.org/abs/2203.06125)
Contributions:14 commits, 3 PRs, 20 pushes in 7 months
graph-neural-networkspre-trainingprotein-representation-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.