Personal Projects

Vislang AI Lab

Currently conducting research in visual language understanding at Rice University, focusing on improving the accuracy and efficiency of vision-language models. This project explores novel approaches to multi-modal learning, investigating how AI systems can better understand and process both visual and textual information simultaneously. Working with state-of-the-art transformer architectures and contributing to advancements in cross-modal attention mechanisms.

King Energy

During my internship at King Energy, I contributed to the development of a sophisticated file management and tagging system. This project involved reengineering data pipelines and integrating Google APIs to streamline file retrieval and categorization, significantly reducing manual processing time and enhancing operational efficiency.

AutoNote

AutoNote is an innovative application that leverages advanced AI technologies to automatically generate comprehensive notes from audio inputs. Designed with students in mind, it utilizes TypeScript and OpenAI Whisper to capture and organize key points from lectures and meetings—allowing users to focus on understanding the material rather than on manual note-taking.

New York City Vehicular Crash Analysis

This project involved analyzing over one million crash records from the NYC Open Data Portal to identify the top ten high-risk factors contributing to vehicular crashes. Utilizing R, SQL, Quarto, and LaTeX, I developed detailed visualizations and deployed a live Shiny web application that delivers data-driven insights and policy recommendations aimed at reducing crash frequency.