Zhen Li is a PhD student in the Department of Computing at The Hong Kong Polytechnic University. He has extensive experience in machine learning systems and AI infrastructure, contributing to the development of advanced AI training engines and inference frameworks, including DLRover, ATorch, Vitis-AI Model Zoo, and Lookahead. Zhen’s work has been applied in production environments, with notable achievements such as Vitis-AI becoming a key deep learning acceleration library in edge computing. Additionally, the ATorch training framework and Lookahead inference algorithm have been deployed at scale in Ant Group’s production, reducing inference costs by 2.6x. His contributions earned him Ant Group’s highest technological honour, the “T-Star” award, in 2024.