Summary
Yiheng Zhang is an Engineering Manager at NVIDIA with nine years of experience building high-performance rendering and deep learning systems, currently leading Inference General Performance efforts for TensorRT. A Stanford MSCS graduate, he moved from hands-on TensorRT development to senior engineering and now management roles, blending low-level optimization with production ML inference. His background spans ray tracing, sparse ML workloads, and parallel rendering across industry and academia, including contributions to MLPerf sparse BERT and TensorRT sparse convolution work. He has taught core computer graphics and vision courses at Stanford and previously led research in rendering denoise and semantic segmentation at Shanghai Jiao Tong University. Outside engineering, Yiheng is a creative polymath who composed, produced, and toured with his own metal band, a detail that reflects his discipline in both technical leadership and creative collaboration.
9 years of coding experience
7 years of employment as a software developer
Bachelor of Science of Engineering - BSE, Computer Science, Bachelor of Science of Engineering - BSE, Computer Science at Shanghai Jiao Tong University
The University of Melbourne
Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at Stanford University
Chinese, English, German