Isaac Yang
Principal Software Engineer at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolIsaac Yang is a Principal Software Engineer based in California with over 20 years of systems and algorithm design experience and 11+ years focused on large-scale deep learning and data pipelines. He combines deep embedded-systems roots (ITRON, uCLinux, Green Hills) with modern ML and big-data practice—spanning C/C++, Python, Matlab, Spark and Hadoop—to deliver production-ready AI in areas from medical imaging to automotive and surveillance. At NVIDIA he architected deep learning data management used across platforms and contributed to high-profile open-source projects such as NVIDIA DIGITS and NVFlare, implementing federated learning validators and model persistence. He has led international teams building cloud document-management and Internet TV subsystems, holds three US patents, and earned 50+ distinction-level MOOCs across AI and ML topics. Known for translating research-grade algorithms into maintainable, high-performance systems, he also brings uncommon breadth: from low-level codec and security stacks in consumer electronics to federated learning validators for distributed model workflows.
11 years of coding experience
28 years of employment as a software developer
Ph.D. Electrical Engineering, Ph.D. Electrical Engineering at University of Southern CaliforniaBS Electrical Engineering, BS Electrical Engineering at National Taiwan University
Chinese, English
Github Skills (12)
machine-learning10
caffe10
deep-learning10
gpu10
python10
federated-learning10
validators10
numpy10
h5py9
back-end-development8
user-interface7
tensorflow4
Programming languages (4)
Github contributions (5)
NVIDIA/NVFlare
Aug 2021 - Jan 2023
NVIDIA Federated Learning Application Runtime Environment
Role in this project:
Back-end Developer
Back-end DeveloperContributions:69 releases, 499 reviews, 148 commits in 1 year 5 months
Contributions summary:Isaac implemented initial versions of the "Hello Numpy - Cross Site Validation" example within the NVIDIA FLARE framework. They added code for saving and registering local models during training using the `MLModelRegistry`. Furthermore, the user wrote the validator responsible for validating models received from the server, retrieving model files and returning validation results. The code contributions center around the core functionalities of federated learning and specifically implementing the trainer, model persistor, and validator components.
pythonnvidiadeep-learningruntimemachine-learning
NVIDIA/DIGITS
Apr 2016 - Nov 2019
Deep Learning GPU Training System
Role in this project:ML Engineer
ML EngineerContributions:4 releases, 87 commits, 76 PRs in 3 years 7 months
Contributions summary:Isaac contributed to the NVIDIA DIGITS project, a deep learning training system. Their work included modifications to support Windows compatibility, addressing issues with the h5py and caffe libraries, and implementing improvements to the user interface by displaying Caffe version information. Furthermore, they made changes to enhance the system's platform support and updated the codebase to handle model downloads and loading, demonstrating involvement in the core functionality and user experience of the deep learning platform. They also optimized the code base.
deep-learninggputorchmachine-learningtraining
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.