Hi, I'm Hongyi (Tom) Chen

I am a Data Science undergraduate at CUHK-Shenzhen and completed a visiting program in Computer Science at UC Berkeley in Fall 2025. My work spans medical AI, large language model research, machine learning engineering, and intelligent data systems.

Hongyi (Tom) Chen

CVs updated Aug 2026

Education

The Chinese University of Hong Kong, Shenzhen

B.S. in Data Science and Big Data Technology

Sep 2023 – Present

University of California, Berkeley

Visiting Student, Computer Science

Aug 2025 – Dec 2025

Experience

Research

CardioFormer: Cardiac Sudden Death Prediction

Apr 2026 – Present

Undergraduate Researcher (Advisor: Prof. Joshua Zhou)

Developing a safe EHR-based decision policy for cardiac sudden death prediction.

SkinGPT-X: Medical Large Language Model

Dec 2025 – Apr 2026

Undergraduate Researcher (Advisor: Prof. Joshua Zhou)

Built a medical LLM evaluation framework and a Self-Evolving Memory module.

Impact of Polysemanticity on Large Language Model Editing

Nov 2025 – Dec 2025

Undergraduate Researcher (Advisor: Prof. Anant Sahai)

Studied collateral damage from neuron-level and layer-level editing in polysemantic LLMs.

Internships

MLE Intern

Jul 2026 – Present

Cardinal Operations, Shanghai, China

Working on machine learning algorithms and applied modeling.

FinTech Intern

Jul 2025 – Aug 2025

China Merchants Bank, Xiamen

Built a BERT-based fraud risk classifier with a 12% relative accuracy gain over XGBoost and used token-level SHAP for model analysis.

Data Analysis Intern

Apr 2025 – Jun 2025

DigitSuper Technology

Improved bid win rate by 20% and ROAS by 10% through advertising-data analysis and A/B tests.

Selected Projects

Browse the public project collection.

View All Projects

Leadership & Extracurricular

Project Lead, Zhixing Social Practice Program

Jun 2024 – Oct 2024

Led survey data processing and feature engineering with Python and Pandas, applied K-Means clustering to identify four core user profiles, and delivered findings adopted by CUHK-Shenzhen for student engagement planning.

Co-Lead, Rural Revitalization Program

Aug 2024

Led stakeholder interviews and primary data collection, structured qualitative findings through thematic analysis, and co-authored a report adopted by the local government and featured by local media.

Languages

Mandarin (Native) English (TOEFL 109)

Technical Skills

Programming & Web

Python R SQL Java C JavaScript HTML/CSS LaTeX Git

Machine Learning & Data

PyTorch Scikit-learn Transformers Pandas NumPy Tableau

Hobbies

  • Soccer
  • Marathon & Hiking
  • Novels & Films
  • Traveling