Jack Zhang
United States
Summary
🤩
Rockstar🎓
Top SchoolJack Zhang is a doctoral student in computer science at Johns Hopkins with eight years of research and engineering experience bridging deep learning systems and NLP. He has made substantive core contributions to PyTorch—improving tensor ops, dynamic-shape performance, and ATen decompositions—which reflect deep familiarity with ML framework internals and GPU-accelerated computation. His work spans high-impact industry research roles at Meta and Microsoft on model alignment, safety, and post-training methods (ICLR paper), alongside internships at MIT CSAIL and JHU on robust and controllable text generation. Comfortable moving between code and theory, Jack combines systems-level optimization skills (C++/PyTorch intern experience) with a strong academic track record and cross-disciplinary projects in speech, social media NLP, and video anonymization. Based in Baltimore, he brings a pragmatic researcher’s mindset to production-relevant ML problems and a proven ability to improve core infrastructure that many downstream models rely on.
8 years of coding experience
4 years of employment as a software developer
University of California, Los AngelesHigh School Diploma, High School Diploma at St. George's School
English, Chinese, Japanese
Github Skills (14)
code-optimization10
pytorch10
machine-learning10
tensor10
deep-learning10
optmization10
optimisation10
python10
optimization10
autograd9
neural-network8
numpy8
gpu8
mlops7
Programming languages (7)
Github contributions (5)
pytorch/pytorch
Aug 2024 - Mar 2025
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
ML Engineer
ML EngineerContributions:9 reviews, 16 PRs, 63 pushes in 7 months
Contributions summary:Jack primarily contributed to the core functionality and optimization of PyTorch, a deep-learning framework. Their work involved refactoring and improving existing functionalities like `index_put` and `masked_select`, which is a critical feature within the framework, resolving issues and improving the efficiency of operations related to dynamic shapes. Furthermore, the user registered decompositions for core aten, which helped to solve the FX verifier and they also worked on STFT support. These contributions demonstrate a deep understanding of PyTorch's internals and are crucial for its performance and compatibility.
pythongpu-accelerationdeep-learninggpunumpy
jackzhxng/executorch
Jun 2024 - Mar 2025
On-device AI across mobile, embedded and edge for PyTorch
Contributions:90 pushes, 21 branches in 9 months
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.