Po-han Huang
Deep Learning Software Engineering Manager at NVIDIA
Champaign, Illinois, Taiwan
Summary
👤
Senior🎓
Top SchoolPo-han Huang is a deep learning software engineering manager at NVIDIA with 10 years of experience accelerating neural network and LLM inference on GPUs through projects like TensorRT, TensorRT-LLM, vLLM, and SGLang. He combines academic research—CVPR‑accepted MS thesis work on multi-view stereopsis—with production-grade CUDA engineering to squeeze maximum performance from diverse model frameworks. At NVIDIA he progressed from hands-on engineer to manager, shipping low-level optimizations that bridge research models and real-world deployment. Based in Champaign and originally from Taiwan, he also focuses on user experience for immersive video in VR, reflecting an unusual cross-section of perception research and systems performance.
10 years of coding experience
6 years of employment as a software developer
Bachelor of Science (B.Sc.) Electrical and Electronics Engineering, Bachelor of Science (B.Sc.) Electrical and Electronics Engineering at National Taiwan UniversityMaster’s Degree Electrical and Electronics Engineering, Master’s Degree Electrical and Electronics Engineering at University of Illinois Urbana-Champaign
Chinese, English, Japanese
Github Skills (2)
pytorch8
deep-learning7
Programming languages (1)
Github contributions (5)
phuang17/phuang17.github.io
Sep 2016 - Nov 2021
Contributions:52 pushes, 3 branches in 5 years 3 months
phuang17/DeepMVS
Feb 2018 - May 2022
DeepMVS: Learning Multi-View Stereopsis
Contributions:11 commits, 3 PRs, 9 pushes in 4 years 3 months
deep-learningpytorchmulti-view-learningmulti-view
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.