Work Experience

Data Analyst Intern, Data Science Foundations Lab, HKUST

Project Overview

Developed a full-stack RAG pipeline to enhance LLM response accuracy and relevance across 100,000+ vector database queries.
Fine-tuned reranking models (Qwen3, FlagReranker) to boost performance on QA benchmarks.
Automated preprocessing and embedding generation pipelines for large-scale datasets.
Explored trade-offs between retrieval accuracy, computational cost, and latency for scalable RAG deployment.

This project deepened my understanding of vector databases, high-dimensional search, and hybrid retrieval methods for LLMs. I focused on bridging optimization and NLP by designing efficient, data-driven retrieval systems.

CS 1110 Teaching Assistant, Cornell University

Role Overview

Led bi-weekly lab sessions to guide students through programming exercises on recursion, object-oriented design, and algorithmic problem-solving.
Provided one-on-one consulting to help students debug code, refine logic, and build conceptual understanding in Python.
Developed instructional materials such as walkthroughs, visual examples, and practice problems to strengthen comprehension.

This experience enhanced my ability to communicate technical concepts, mentor effectively, and foster analytical thinking — skills that directly support data-driven research and technical collaboration.

AEW Facilitator, College of Engineering, Cornell University

Workshop Overview

Designed interactive workshops and problem sets to reinforce key concepts in Multivariable Calculus.
Encouraged collaborative learning by guiding peer discussions and supporting diverse problem-solving strategies.
Clarified challenging topics through examples connecting theory to practical applications in engineering and data analysis.

Through leading AEW sessions, I developed strong communication, facilitation, and problem-solving skills while helping students build confidence and technical proficiency.

Academic Excellence Workshop demonstration