CV

You can download this resume here (PDF).

Education

• Ph.D. in Applied Mathematics and Computational Science, University of Pennsylvania, May 2021 – December 2026 (expected)
• M.A. in Applied Mathematics and Computational Science, University of Pennsylvania, August 2019 – May 2021
  • Cumulative GPA: 4.00/4.00
• B.S. in Mathematics with Honor / B.A. in Chemistry, Wake Forest University, August 2015 – May 2019
  • Cumulative GPA: 3.89/4.00, Summa Cum Laude
  • Minor in Computer Science

Work Experience

• Summer Engineering Associate, Risk Economics Strats, Goldman Sachs Group, Inc., June 2025 – August 2025
  • New York, NY
  • Developed an automated credit rating system using LangChain and LLMs to extract and evaluate qualitative criteria from credit review documents, generating structured outputs with full references and justifications for human review.
• Summer Engineering Associate, Market Risk Strats, Goldman Sachs Group, Inc., June 2024 – August 2024
  • New York, NY
  • Projects: Large Language Model uncertainty quantification, regression test for market risk scenarios, PnL hedging for desk portfolios.
  • Explored quantitative and qualitative ways to evaluate the effectiveness and confidence of Generative AI solutions in ESG investing.
  • Created and maintained programs for market risk evaluation.
• Graduate Research Assistant, University of Pennsylvania, Penn Institute of Computational Science, May 2020 – Present
  • Philadelphia, PA
  • Projects: Autoregressive prediction of time-dependent PDEs, Physics-Informed DeepONet learning the solution operator of parametric PDEs, analysis of eigenvector bias of Fourier feature networks on PINNs, COVID-19 dynamics modeling with LSTM and NeuralODE, segmentation of platelets from hemostatic plug scans.
  • Implemented training, evaluation, and inference pipelines for SOTA computer vision, computer graphics, and scientific machine learning models.
  • Curated non-standard datasets for disease transmission and biomedical imaging projects.

Skills

• Programming Languages: Python, R, MATLAB, LaTeX
• Machine Learning Packages: JAX, Optax, Flax, LangChain, PyTorch, Scikit-learn
• Languages: English, Chinese

Publications

Enhancing the trainability and expressivity of deep MLPs with globally orthogonal initialization

On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networks

Learning the solution operator of parametric partial differential equations with physics-informed DeepONets