Zaifeng Pan
Scholar

Zaifeng Pan

Google Scholar ID: Ryyb18EAAAAJ
University of California, San Diego
Machine Learning Systems
Citations & Impact
All-time
Citations
123
 
H-index
5
 
i10-index
5
 
Publications
11
 
Co-authors
3
list available
Resume (English only)
Academic Achievements
  • - NeurIPS’25: KVFlow: Efficient Prefix Caching for Accelerating LLM-Based Multi-Agent Workflows
  • - MLSys’25: FastTree: Optimizing Attention Kernel and Runtime for Tree-Structured LLM Inference
  • - SC’24: RecFlex: Enabling Feature Heterogeneity-Aware Optimization for Deep Recommendation Models with Flexible Schedules
  • - ASPLOS’23: RECom: A Compiler Approach to Accelerating Recommendation Model Inference with Massive Embedding Columns
Research Experience
  • - Amazon Web Service, Applied Scientist Intern, 2025, Mentor: Dr. Aninda Manocha and Dr. Zhuang Wang
  • - Microsoft, Research Intern, 2023 - 2024, Mentor: Dr. Zhen Zheng
  • - Alibaba PAI, Research Intern, 2021 - 2023, Mentor: Dr. Zhen Zheng
Education
  • - Ph.D.: University of California, San Diego (UCSD), Computer Science & Engineering, 2024 - Present, Advisor: Prof. Yufei Ding
  • - M.S.: Renmin University of China (RUC), Computer Software and Theory, 2021 - 2024, Advisor: Prof. Feng Zhang
  • - B.E.: Shanghai Jiao Tong University (SJTU), Mechanical Engineering, 2017 - 2021
Background
  • Research Interests: Machine learning systems, with a particular focus on making large language models (LLMs) more efficient; Professional Field: Systems, compilers, and kernels; Background: Currently a Ph.D. student at UC San Diego, previously earned a Master’s degree from Renmin University of China.
Miscellany
  • Chinese Name: 潘再峰; Nickname: 小再; Contact: zapan@ucsd.edu; Roommate is an expert in computer architecture, check out his homepage for more information.