Summary
Haojie Liu is a Senior Architect based in Shanghai with nine years of experience at the intersection of deep learning and video systems, currently driving RTX Video Super Resolution productization and cloud/hybrid AI video coding at NVIDIA. He holds a Ph.D. from Nanjing University with a research focus on end-to-end image and video compression and spent time as a visiting scholar at NYU working on E2E video coding. Haojie blends research rigor with productization experience—shipping DL-based deblurring, artifact removal, and real-time video enhancements for production pipelines. He coordinated the release of end-to-end compression work from his research, signaling both leadership in academic projects and practical open deployment. Comfortable moving between model research and system integration, he brings deep domain expertise in video coding, machine learning, and low-level media processing.
9 years of coding experience
Doctor of Philosophy - PhD, Video Coding, Machine Learning, Video Processing, Doctor of Philosophy - PhD, Video Coding, Machine Learning, Video Processing at Nanjing University
Visiting Scholar, End-to-End Video Coding and Video Processing, Visiting Scholar, End-to-End Video Coding and Video Processing at New York University