Lifu Zhang
Senior Deep Learning Architect at NVIDIA
San Francisco Bay Area United States
Summary
🤩
Rockstar🎓
Top SchoolLifu Zhang is a Senior Deep Learning Architect at NVIDIA with five years of focused experience building high-performance neural network training systems and production-ready ML infrastructure. He holds an M.A.Sc. from the University of Toronto where he developed a pipelined DNN training method that improved ResNet training speed by 1.82X while halving memory footprint, and published the results. At NVIDIA he progressed from architect to senior architect, applying deep expertise in PyTorch, TensorFlow, CUDA and C/C++ to scale training and inference across GPU platforms. His background in hardware engineering and FPGA timing tools gives him a rare cross-layer perspective on optimizing ML models from silicon to software. Colleagues describe him as research-minded but pragmatic—able to translate novel algorithms into robust, deployable systems in the San Francisco Bay Area.
5 years of coding experience
5 years of employment as a software developer
High School, High School at York Mills CIBachelor of Applied Science - BASc, Engineering Science, Electrical and Computer Option, Bachelor of Applied Science - BASc, Engineering Science, Electrical and Computer Option at University of Toronto
English, Chinese
Github Skills (35)
speaker-diarization10
speech-to-text10
machine-translation10
python10
machine-learning10
speech-synthesis10
multimodal10
text-normalization10
large-language-models10
nmt10
deep-learning10
gpu10
generative-ai10
speaker-recognition10
natural-language-processing10
Programming languages (1)
Github contributions (5)
tomlifu/NeMo
Jun 2024 - Aug 2024
A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)
Contributions:10 pushes, 2 branches in 1 month
tomlifu/Megatron-LM
Oct 2024 - Mar 2025
Contributions:3 pushes, 1 branch in 5 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.