My image.


Hi, my name is Zhenyu Song (宋振宇, pronounced as zen-u). I am an Applied Scientist at AWS AI Labs, working on systems for large language models.

I obtained my Ph.D. degree from Princeton CS Department, under the supervision of Prof. Kai Li and Prof. Wyatt Lloyd, where I worked on using machine learning to improve content delivery network (CDN) caching.

Email: zhenyus [at]


  • HALP: Heuristic Aided Learned Preference Eviction Policy for YouTube Content Delivery Network. [PDF] [Video]
    Zhenyu Song, Kevin Chen, Nikhil Sarda, Deniz Altınbüken, Eugene Brevdo, Jimmy Coleman, Xiao Ju, Pawel Jurczyk, Richard Schooler, Ramki Gummadi.
    NSDI 2023

  • Learning Relaxed Belady for Content Distribution Network Caching. [PDF] [Code] [Dataset] [Video]
    Zhenyu Song, Daniel S. Berger, Kai Li, Wyatt Lloyd.
    NSDI 2020

  • Wi-Fi Goes to Town: Rapid Picocell Switching for Wireless Transit Networks. [PDF] [Slides] [Video] [Demo]
    Zhenyu Song, Longfei Shangguan, Kyle Jamieson.
    SIGCOMM 2017

  • Modeling Topic-level Academic Influence in Scientific Literatures. [PDF][Slides]
    Jiaming Shen, Zhenyu Song, Shitao Li, Zhaowei Tan, Yuning Mao, Luoyi Fu, Li Song, Xinbing Wang.
    AAAI 2016 Workshop on Scholarly Big Data

  • HiQuadLoc: A RSS Fingerprinting based Indoor Localization System for Quadrotors. [PDF]
    Xiaohua Tian, Zhenyu Song, Binyao Jiang, Yang Zhang, Tuo Yu, Xinbing Wang.
    IEEE TMC 2016


My research has been made possible through collaborations with several incredible mentors and mentees.



  • Maxwell Xu (Princeton Undergrad 2021) → Microsoft
  • Audrey Cheng (Princeton Undergrad 2020) → Berkeley PhD Program
  • John Suh (Princeton Undergrad 2020) → Robinhood


External Reviewer

  • OSDI 2022
  • OSDI 2018