Portrait of Mingkai Li

Mingkai Li

Ph.D. Student

Columbia University mingkai.li@columbia.edu

About Me

I am a Ph.D. student at Columbia University, advised by Prof. Tanvir Ahmed Khan, and working closely with Prof. Suman Jana and Prof. Joe Devietti.

My research focuses on building reliable and secure infrastructure for autonomous AI systems across the hardware-software stack. I am particularly interested in understanding and addressing security vulnerabilities and failures, from accelerators and GPU kernels to LLM infrastructure and emerging agentic systems.

Prior to Columbia, I received my B.E. in Computer Science from University of Science and Technology of China (USTC). I have had the pleasure of working with Prof. Prateek Saxena and Prof. Trevor E. Carlson at National University of Singapore, and Prof. Xianghang Mi at USTC.

Outside of research, I enjoy cycling, squash, and tennis.

Feel free to reach out if you share similar interests!

News

Publications

  1. Teaser image for NanoTag: Systems Support for Efficient Byte-Granular Overflow Detection on ARM MTE Oakland 26
    NanoTag: Systems Support for Efficient Byte-Granular Overflow Detection on ARM MTE
    Mingkai Li, Hang Ye, Joseph Devietti, Suman Jana, Tanvir Ahmed Khan
    The 47th IEEE Symposium on Security and Privacy (S&P), 2026.
    PDF Code Slides Poster BibTeX
  2. Teaser image for Caplification: Bridging Capability-Aware and Capability-Oblivious Software SACMAT 25
    Caplification: Bridging Capability-Aware and Capability-Oblivious Software
    Jason Zhijingcheng Yu, Mingkai Li, Aditya Badole, Trevor E. Carlson, Michael Swift, Prateek Saxena
    The 30th ACM Symposium on Access Control Models and Technologies (SACMAT), 2025.
    PDF Code Project Page BibTeX
  3. Teaser image for TransMap: Pinpointing Mistakes in Neural Code Translation FSE 23
    TransMap: Pinpointing Mistakes in Neural Code Translation
    Bo Wang, Ruishi Li, Mingkai Li, Prateek Saxena
    The 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE), 2023.
    PDF Code Project Page BibTeX

Workshops

  1. Teaser image for Challenges and Design Considerations for Finding CUDA Bugs Through GPU-Native Fuzzing HotEthics 26
    Challenges and Design Considerations for Finding CUDA Bugs Through GPU-Native Fuzzing
    Mingkai Li, Joseph Devietti, Suman Jana, Tanvir Ahmed Khan
    The 2nd Workshop on Ethical Systems and Architecture Design (HotEthics), 2026 (co-located with ASPLOS '26).
    PDF Slides BibTeX

Talks

Teaching