Back to homepage

Ruipu Li

Ph.D. Student in Computer Science · University of Michigan

Education

University of Michigan, Ann Arbor

Ph.D. Student, Computer Science

Aug 2024 – Present

University of Michigan, Ann Arbor

M.S., Computer Science

Aug 2022 – Apr 2024

University of Michigan, Ann Arbor

B.S., Computer Science

Aug 2020 – Apr 2022

Shanghai Jiao Tong University

B.S., Electrical and Computer Engineering

Aug 2018 – Aug 2022

Publications

Neural Conformal Control for Time Series Forecasting

Ruipu Li and Alexander Rodríguez.

Proceedings of the 39th AAAI Conference on Artificial Intelligence (AAAI ’25), 2025.

Optimization-based Online Conformal Prediction for Multi-step Forecasting

Ruipu Li, Diego Menacho, and Alexander Rodríguez.

Submitted to the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’26), 2026.

Research projects

CausalSim: Generative Causal AI for Strategic Decision Support

Graduate Researcher, University of Michigan

Jan 2026 – Present

  • Build an LLM-driven reasoning pipeline combining retrieval with structural causal models to debug discrepancies across black-box simulators.
  • Engineer infrastructure to ground explanations in domain constraints for trustworthy what-if analysis.

ML to Reduce Uncertainty in Climate Forcing by Aerosols

Research Assistant, Climate and Space Sciences and Engineering, University of Michigan

Aug 2022 – Aug 2024

  • Process satellite observations using interpolation, PCA, and feature analysis for aerosol-cloud modeling inputs.
  • Train predictive models for aerosol optical depth and cloud droplet number concentration.

Collaborations and applied forecasting

FluSight Forecasting Challenge (CDC)

Team Lead, UM-DeepOutBreak, University of Michigan

Jul 2023 – Present

  • Top-3 accuracy (MAE) out of about 40 teams in the 2023–2024 FluSight season.
  • Built an end-to-end probabilistic forecasting pipeline for CDC ensemble submissions.
  • Develop calibrated multi-step prediction intervals using conformal prediction methods.

Reliable Uncertainty Cones for Cyclone Trajectory Prediction

Collaboration with Google DeepMind WeatherLab

Oct 2025 – Present

  • Apply an adaptive conformal method (CP-Traj) to ensemble cyclone forecasts for long-horizon coverage.
  • Coordinate larger-ensemble experiments and possible integration into trajectory uncertainty products.

Teaching

Graduate Student Instructor (GSI), EECS 492

Introduction to Artificial Intelligence

Jan 2026 – Present

  • Design and release weekly homework aligned with course learning goals.
  • Lead discussion sections and support students in problem solving and core AI concepts.

Industry

NIO, Autonomous Driving System

Machine Learning Intern

May 2021 – Aug 2021

  • Curated and cleaned large-scale road-test logs to build datasets for lane-change modeling.
  • Improved lane-change decision-making with an LSTM-based motion prediction model.
  • Built Python tooling for road-test analysis and integrated automated testing.

Honors

  • Graduated summa cum laude, University of Michigan
  • Dean’s List, University of Michigan
  • Undergraduate Excellent Scholarship, Shanghai Jiao Tong University
  • John Wu and Jane Sun Excellence Scholarship (Top 3%)

Skills

  • Programming: C, C++, C#, Python, JavaScript, Go, Rust
  • Systems and tools: Great Lakes HPC, Google Cloud Platform (GCP)

Hobbies

Badminton Street photography Table tennis Skiing