Xinyu Fei

About Me

Xinyu Fei is an Applied Scientist II at Amazon.com, focusing on ML-integrated mixed-integer optimization algorithm design and GenAI reasoning for optimization. Xinyu received her Ph.D. in Operations Research and Scientific Computing from the University of Michigan, advised by Prof. Siqian Shen.

Xinyu has been awarded:

• Pritsker Doctoral Dissertation Award, IISE 2024
• Rackham Travel Grant, UM 2022, 2023
• Michigan Institute for Computational Discovery and Engineering (MICDE) Student Fellowship, UM 2021
• Engineering Graduate Fellowship, UM 2019

Xinyu’s Resume

Research Interests

• Optimization: Mixed-integer optimization, stochastic optimization, large-scale optimization
• Machine Learning: ML-integrated optimization, GenAI agents, reinforcement learning (ongoing research)
• Applications: Quantum computing, traffic optimization, resource planning, labor optimization

Technical Skills

• Programming Languages: Python, Java, Pytorch, C++, SQL, Matlab, R
• Solvers & Tools: Xpress, Gurobi, AWS (S3, SageMaker, Lambda)

Education

• Sept. 2019 - Dec. 2023
  Ph.D. in Operations Research and Scientific Computing
  University of Michigan
  Focus: Mixed-integer & stochastic optimization, reinforcement learning
  Advisor: Prof. Siqian Shen

• Sept. 2015 - Jun. 2019
  B.S. in Computational and Applied Mathematics
  School of the Gifted Young (SCGY)
  University of Science and Technology of China

Work Experience

• Feb 2024 – Present
  Applied Scientist II
  Amazon.com
  • Scalable Optimization System for Real-Time Labor Planning
  • AI Reasoning Model for Explainable Optimization

• May 2023 – Aug 2023
  Research Scientist Intern
  Amazon.com
  • Machine Learning Based Ergonomics Stowing Assistance

• May 2022 – Aug 2022
  Research Scientist Intern
  Amazon.com
  • Decomposition Algorithm for Large-scale Resource Planning Problems

• May 2021 – Feb 2024
  Research Collaborator
  Argonne National Laboratory
  • Binary control for quantum systems under uncertainty

Publications

• Binary Quantum Control Optimization with Uncertain Hamiltonians
  Xinyu Fei , Lucas T. Brady , Jeffrey Larson , Sven Leyffer , Siqian Shen
  INFORMS Journal on Computing, 2025
  [Paper]
• Switching Time Optimization for Binary Quantum Optimal Control
  Xinyu Fei , Lucas T. Brady , Jeffrey Larson , Sven Leyffer , Siqian Shen
  ACM Transactions on Quantum Computing, 2025
  [Paper]
• Traffic Signal Control under Stochastic Traffic Demand and Vehicle Turning via Decentralized Decomposition Approaches
  Xinyu Fei , Xingmin Wang , Xian Yu , Yiheng Feng , Henry Liu , Siqian Shen , Yafeng Yin
  European Journal of Operational Research, 2023
  [Paper]
• Binary Control Pulse Optimization for Quantum Systems
  Xinyu Fei , Lucas T. Brady , Jeffrey Larson , Sven Leyffer , Siqian Shen
  Quantum, 2023
  [Paper]
• An Optimization-and-Simulation Framework for Redesigning University Campus Bus System with Social Distancing
  Gongyu Chen , Xinyu Fei , Huiwen Jia , Xian Yu , Siqian Shen
  INFORMS Journal on Applied Analytics, 2023
  [Paper]
• huge: High-dimensional Undirected Graph Estimation
  Haoming Jiang , Xinyu Fei , Han Liu , Kathryn Roeder , John Lafferty , Larry Wasserman , Xingguo Li , Tuo Zhao
  R package version, 2019
  [Paper]

Presentations & Talks

• Data-Driven Automated Labor Staffing by Real-time Optimization in Amazon Fulfillment Centers
  INFORMS Annual Meeting, Seattle, October 2024
• Route Recommendation Optimization under Unknown and Uncertain Travelers’ Trust
  INFORMS Annual Meeting, Phoenix, October 2023
• Binary Control Pulse Optimization for Quantum Systems
  INFORMS Annual Meeting, Indianapolis, October 2022
• Quantum Control Pulse Optimization With Uncertain Hamiltonian Controllers
  INFORMS Annual Meeting, Indianapolis, October 2022
• Optimization Methods for Discrete Binary Quantum Control
  Workshop on Optimization and Machine Learning, Evanston, May 2022
• Optimizing Reopening and Lockdown of Correlated Businesses and Regions under Uncertain Disease Outbreaks
  IISE Annual Meeting, Virtual, May 2021
• Distributed Optimization of Traffic Signals in Arterial Networks
  INFORMS Annual Meeting, Virtual, November 2020