I am a Senior Applied Scientist at Amazon Annapurna Labs. I currently work on AWS Neuron and develop optimizations to make state-of-the-art LLM models execute efficiently on AWS Trainium Chips.

I received my Ph.D. at New York University. I was advised by Professor Jinyang Li. My research focused on improving the execution performance of Deep Learning serving systems, particularly for Large Language Models.

Before joining NYU, I received my B.S. in Computer Science at Nanjing University, where I worked on improving performance of wireless network and was advised by Professor Wenzhong Li.

Here is my resume.

Education

• Ph.D. in Computer Science
  New York University, New York, NY, United States

• B.S. in Computer Science
  Nanjing University, Nanjing, Jiangsu, China

Selected Publications

• CRAFT: Fine-Grained Cost-Aware Expert Replication For Efficient Mixture-of-Experts Serving [paper]
  Adrian Zhao, Zhenkun Cai, Zhenyu Song, Lingfan Yu, Haozheng Fan, Jun Wu, Yida Wang, Nandita Vijaykumar
  MLSys’26
  Bellevue, USA, May 2026
• SAS: Sparse Attention Synthesizer for Efficient Language Model Inference [paper]
  Yuan Zhou, Shaojie Xiang, Lingfan Yu, Zhenyu Song, Charith Mendis, Yida Wang
  Eurosys’26
  Edinburgh, Scotland UK, April 2026
• Stateful Large Language Model Serving with Pensieve [paper]
  Lingfan Yu, Jinkun Lin, Jinyang Li
  Eurosys’25
  Rotterdam, Netherlands, March 2025
• Scalable Graph Neural Networks for Heterogeneous Graphs [paper]
  Lingfan Yu, Jiajun Shen, Jinyang Li, Adam Lerer
  ArXiv preprint arXiv:2011.09679
• Deep Graph Library: Towards Efficient and Scalable Deep Learning on Graphs [paper]
  Minjie Wang, Lingfan Yu, Da Zheng, Quan Gan, Yu Gai, Zihao Ye, Mufei Li, Jinjing Zhou, Qi Huang, Chao Ma, Ziyue Huang, Qipeng Guo, Hao Zhang, Haibin Lin, Junbo Zhao, Jinyang Li, Alexander Smola and Zheng Zhang
  ICLR’19 Workshop on Representation Learning on Graphs and Manifolds
  New Orleans, USA, May 2019
• Low Latency RNN Inference with Cellular Batching [paper]
  Pin Gao, Lingfan Yu (co-first author), Yongwei Wu, Jinyang Li.
  Eurosys’18
  Porto, Portugal, April 2018.
• The Efficient Server Audit Problem, Deduplicated Re-execution, and the Web [paper]
  Cheng Tan, Lingfan Yu, Joshua B. Leners, Michael Walfish.
  SOSP’17 (Best paper award)
  Shanghai, China, October 2017

Work Experience

• Amazon Annapurna Labs
  2024.9 - Present
  Senior Applied Scientist

• Facebook, Inc. Facebook AI Research (FAIR)
  2020.5 - 2020.10
  Research Intern

• Amazon Web Services, Inc.
  2019.6 - 2019.8
  Applied Scientist Intern