About

Current

I am currently working as an AI Research & Application Engineer at Sapient Intelligence, applying Hierarchical Reasoning Models to healthcare scenarios and collaborating with Peking Union Medical College Hospital to develop healthcare solutions.

Research

At Shanghai Jiao Tong University, I developed PLASMA, a module based on optimal transport for protein substructure alignment (accepted at ICLR 2026), supervised by Prof. Pietro Lio and Prof. Liang Hong.

At Cambridge, I worked on hierarchical protein representation learning under Prof. Pietro Lio, developing TCPNet — an SE(3)-equivariant topological neural network that simultaneously learns protein residue-level and secondary structure-level features (Invited Talk). I also worked on multi-omics integration with graph neural networks and graph-based facial expression analysis under Prof. Hatice Gunes, resulting in the GraphAU-Pain paper presented at IJCAI25.

At UCL, I collaborated with clinicians from UCL Hospital on readmission prediction using remote patient monitoring data, building interpretable ML models with SHAP analysis under Prof. Ivana Drobnjak.

Education

At the University of Cambridge, I completed my M.Phil. in Advanced Computer Science with Distinction (ranked 8th out of 60). My dissertation Topotein: Topological Deep Learning for Protein Representation Learning, supervised by Prof. Pietro Lio, received a score of 88/100 and won the Highly Commended M.Phil Project 2024-2025 prize. Courses included Geometric Deep Learning, Affective AI, Mobile & Wearable ML, NLP, and Multi-agent RL.

I previously graduated with First-Class Honors in B.Sc. Computer Science from University College London. Courses included Machine Learning, Computer Vision, Reinforcement Learning, Mathematics & Statistics, and Software Engineering.

Industry Experience

At Xunfei Healthcare Technology I built a knowledge distillation pipeline using a teacher LLM to synthesize medical instruction examples, improving student model (Xiaoyi) instruction-following accuracy from 34% to 82% via supervised fine-tuning.

At Luojin Data Information I developed a Retrieval-Augmented Generation (RAG) pipeline with MongoDB vector store and LangChain agents for automated financial report classification, achieving 92% query translation accuracy and 97% entity matching rate.

I also collaborated with IBM and the NHS to create an automated chatbot generation service for hospitals, reducing development time from weeks to 30 minutes. The project was demonstrated live at Great Ormond Street Hospital and presented to IBM, Microsoft, and Intel.

Research Interests

I am interested in geometric deep learning, large language models, and generative models. I’m dedicated to creating novel ML methods that address pressing biomedical and healthcare challenges. I welcome opportunities to collaborate on innovative, high-impact projects!

Technical Skills

Machine Learning: PyTorch, NumPy, Polars

Visualization: Matplotlib, Seaborn, Plotly