I am a 1st year Ph.D. student at the University of Michigan, advised by Prof. Ang Chen. I am exploring AI agents and their applications, specializing in AI-driven solutions for cloud management.

Previously, I completed my M.S. in CSE at the same institution, where I worked with Prof. Mosharaf Chowdhury on DNN training. and with Prof. Xianglei Huang applying ML techniques to atmospheric research (PREFIRE, NASA).

Email / CV / Google Scholar / GitHub / LinkedIn / Twitter

News

• [12-09-2024] TerraFault Accepted to AIOps’25!

• [06-04-2024] 'PREFIRE and Ice’ Launch – the second of two launches for NASA's PREFIRE mission! YouTube

• [05-25-2024] 'Ready, Aim, PREFIRE’ Launch – the first of two launches for NASA's PREFIRE mission! YouTube

• [05-17-2024] Two PREFIRE CubeSats are scheduled for launch in May 2024.

• [05-04-2024] Graduated from the University of Michigan with a M.S. in Computer Science!

• [07-14-2023] Oobleck Accepted to SOSP’23!

• [07-12-2023] SmarCyPad Accepted to IMWUT’23!

• [04-03-2023] One paper Accepted to IGARSS’23!

• [03-04-2023] Chase Accepted to ICLR’23 workshop!

• [11-10-2022] Carbon-Aware Zeus Wins Second Place at CarbonHack’22.

• [05-21-2022] Graduated from the University of Tennessee!

Publications

• Automated Bug Discovery in Cloud Infrastructure-as-Code Updates with LLM Agents
  Yiming Xiang*, Zhenning Yang*, Jingjia Peng, Hermann Bauer, Patrick Tser Jern Kon, Yiming Qiu, Ang Chen
  AIOps 2025 / To Appear

• Oobleck: Resilient Distributed Training of Large Models Using Pipeline Templates
  Insu Jang, Zhenning Yang, Zhen Zhang, Xin Jin, Mosharaf Chowdhury
  SOSP 2023 / Paper / Code

• SmarCyPad: A Smart Seat Pad for Cycling Fitness Tracking Leveraging Low-cost Conductive Fabric Sensors
  Yi Wu, Luis González Villalobos, Zhenning Yang, Gregory Thomas Croisdale, Çağdaş KARATAŞ, Jian Liu
  IMWUT 2023 / Paper / Video

• A Fast Neural Network-Based Approach for Joint Mid-IR and Far-IR Surface Spectral Emissivity Retrieval
  Zhenning Yang, Xiuhong Chen, Xianglei Huang, Tristan L’Ecuyer, Brian Drouin
  IGARSS 2023 / Paper / Poster

• Chasing Low-carbon Electricity for Practical and Sustainable DNN Training
  Zhenning Yang, Luoxi Meng, Jae-Won Chung, Mosharaf Chowdhury
  ICLR Workshop: Tackling Climate Change with Machine Learning, 2023
  Paper / Code / Poster / Video

Other Projects

• Curie: Toward Rigorous and Automated Scientific Experimentation with AI Agents
  Patrick Tser Jern Kon, Jiachen Liu, Qiuyi Ding, Yiming Qiu, Zhenning Yang, Yibo Huang, Jayanth Srinivasa, Myungjin Lee, Mosharaf Chowdhury, Ang Chen
  arXiv 2025 / Paper

• Disaggregating Embedding Recommendation Systems with FlexEMR
  Yibo Huang, Zhenning Yang, Jiarong Xing, Yi Dai, Yiming Qiu, Dingming Wu, Fan Lai, Ang Chen
  arXiv 2024 / Paper

• Aether: Efficient Serving of Large-scale Vector Search with Sharded Indexes
  Zhenning Yang, Jeremy Flics, Jiaheng Lu
  In 2024 / Paper / Code / Poster

• SLRNet: Semi-Supervised Semantic Segmentation Via Label Reuse for Human Decomposition Images
  Sara Mousavi, Zhenning Yang, Kelley Cross, Dawnie Steadman, Audris Mockus
  arXiv 2022 / Paper / Code

• Pseudo Pixel-level Labeling for Images with Evolving Content
  Sara Mousavi, Zhenning Yang, Kelley Cross, Dawnie Steadman, Audris Mockus
  arXiv 2021 / Paper

Teaching

• Research Experiences for Undergraduates (REU) Student Mentor, Summer 2020

• Teaching Assistant, MATH 123: Finite Mathematics, Summer 2019

• Math Tutor, Fall 2018 – Summer 2019

Media Coverage

• [2023] UM-CSE News: Open-source training framework increases the speed of large language model pre-training when failures arise

• [2023] UM-CSE News: CSE researchers present new findings at OOPSLA and SOSP

• [2022] UM-CSE News: Researchers cut down on AI’s carbon footprint with new optimization framework

• [2022] TechTarget: Microsoft offers developer guidance on green software tools

• [2022] UM News: Optimization could cut the carbon footprint of AI training by up to 75%

• [2022] The Michigan Daily: Researchers at UMich find a way to reduce the energy consumed by AI

• [2022] Linkedin: Linkedin post from Benoît de Chateauvieux