Shengliang Xu is a senior deep learning engineer on the NVIDIA Algorithmic Model Optimization team focused on end-to-end optimization of deep learning model inference on NVIDIA GPU platforms. His research and development interests span model and inference system optimization of large language models and large generative models. Shengliang holds an M.S. degree in computer science from University of Washington, where he dropped out the Ph.D. program. He holds another M.S. degree and a B.S. degree both in computer science from Shanghai Jiao Tong University.