Experience

Research

Pure Nash Equilibria under the Affine Mechanism

University of Wisconsin–Madison

  • First author; paper accepted for oral presentation at GameSec 2026.
  • Characterized the Nash equilibrium of the Affine Mechanism Game under complete information.
  • Extended the analysis to the incomplete-information setting and characterized its Bayesian Nash equilibrium.
  • Co-authored the manuscript and drafted most of its formal proofs.
  • Developed Python programs to compute and visualize game behavior across multiple scenarios.

Partially Observable Markov Game Solver

University of Wisconsin–Madison

  • Implemented Nash Q-learning and deep Q-learning methods in Python and PyTorch for partially observable Markov game environments.
  • Analyzed algorithmic solvability across game scenarios; identified solvable cases and diagnosed causes of non-convergence.

Sturgeon Re-Identification

Beijing Academy of Agriculture and Forestry Sciences

  • Contributed to a deep learning pipeline for sturgeon re-identification (ReID).
  • Curated a 6,000-image dataset and designed image acquisition and labeling protocols.
  • Fine-tuned and integrated YOLOv8 as the detection module in the pipeline; generated detector-aligned crops for ReID training and inference.

Professional

Software Development Intern

Epic Systems

  • Designed and implemented an automated workflow for Beaker, Epic's Lab Information System.
  • Built frontend workflow logic in JavaScript and C# and developed a backend prototype for transforming unstructured input into structured records.
  • Authored design documents and led feedback sessions with cross-functional teams and customers to validate requirements and refine the workflow.

Peer Mentor

University of Wisconsin–Madison · CS 540: Introduction to AI

  • Hosted weekly office hours, helping students understand AI concepts, assignments, and debugging strategies.
  • Reported student feedback and assignment issues to instructors.

Software Engineering Intern

Bohr Systems

  • Designed a camera-only visual localization pipeline for UAVs operating in GPS-denied environments.
  • Evaluated localization algorithms and recommended ORB based on system requirements.
  • Implemented an end-to-end ORB localization prototype in Python, including feature extraction, matching, geometric verification, and map search.