个人简介

JK的研究主要关于面向AI系统的软硬件协同设计 (e.g., RhymeRL, FlowANN)，以及高性能操作系统（OS内核, e.g., Copier）。指导老师：杜东, 夏虞斌, 陈海波。

英文主页：https://he-jingkai.github.io

教育经历

• 2019.09 - 2023.6，上海交通大学软件学院，本科生
• 2023.9 - 2029.3，上海交通大学人工智能学院，博士生

研究方向

Hardware-Software Co-design, OS, MLsys

论文

• [SOSP’26]. Jingkai He, Guangda Sun, Tianjian Li, Dong Du, Yubin Xia, Haibo Chen. Taming Dynamism on GPUs: Cross-SM Kernel Fusion via SM Cooperation and Just-in-Time Reduction. Proceedings of the 32nd ACM Symposium on Operating Systems Principles (SOSP’26). September, 2026. paper

• [OSDI’26]. Haoru Zhao*, Jingkai He*, Jingyao Zeng, Mingkai Dong, Dong Du (* co-first author). Disentangling Graph Dependencies for Efficient Billion-Scale GPU Vector Search. 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI’26). July, 2026.

• [ASPLOS’26]. Jingkai He, Tianjian Li, Erhu Feng, Dong Du, Qian Liu, Tao Liu, Yubin Xia, Haibo Chen. History Doesn’t Repeat Itself but Rollouts Rhyme: Accelerating Reinforcement Learning with RhymeRL. Proceedings of the 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS’26). March, 2026. paper

• [SOSP’25 Best Paper]. Jingkai He, Yunpeng Dong, Dong Du, Mo Zou, Zhitai Yu, Yuxin Ren, Ning Jia, Yubin Xia, Haibo Chen. How to Copy Memory? Coordinated Asynchronous Copy as a First-Class OS Service. Proceedings of the 31st ACM Symposium on Operating Systems Principles (SOSP’25). October, 2025. paper