Dr. Kun Li is currently a Senior Researcher at Microsoft Research, specializing in high-performance computing (HPC) systems. He received his Ph.D. degree in Computer Architecture from the State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS) in 2022.

He has been recognized with numerous honors, including the CCF Outstanding Doctoral Dissertation Award, ACM SIGHPC China Outstanding Doctoral Dissertation Award, CCF Youth Talent Award in HPC, and ACM SIGHPC China Rising Star Award. His work has been published in top-tier CCF-A conferences such as SC, PPoPP, ATC, ASPLOS, and ICS, earning the Best Paper Award at PPoPP’24, SC’25 Best Student Paper Award Finalist, and SC’25 Reproducibility Challenge Candidate.

He serves as a keynote speaker at CCF HPC China 2024, as an Executive Committee Member of the CCF Technical Committee on High-Performance Computing and Computer Architecture, and as a Program Committee Member for PPoPP 2026.

My research focuses on highly scalable computing systems at the intersection of high-performance computing (HPC) and artificial intelligence (AI). I aim to advance this vision through three core research directions:

• Scalable foundation models – Enhancing the efficiency and scalability of large-scale model training and inference on cutting-edge HPC infrastructures.
• Matrix-centric scientific computing – Transforming scientific algorithms and applications into unified matrix-based representations to break traditional trade-offs in performance, accuracy, and scalability.
• Embodied scientific intelligence – Advancing reinforcement learning and AI4Science to create autonomous agents capable of multi-scale scientific reasoning and discovery.