Conducting research at the MIT CCI lab alongside a Professor, focusing on The Deliberatorium, a platform that supports structured, large-scale discussions. We are researching how to unbundle posts that include multiple points—such as various arguments or answers—into single, focused entries. This structure is essential for better organization, analysis, and summarization of ideas on the platform.

Conducting research at the Stanford Virtual Human Interaction Lab under a Postdoctoral researcher. Investigating how different communication methods, including virtual reality, enhance learning in environmental science, with a focus on climate change education

- Fully paid merit based scholarship - Conducting research on expanding AAVENUE's benchmark. Multi-AAVENUE, a larger and more comprehensive benchmark focused on evaluating language models across different dialects

- Conducted research under UC Berkeley and Cornell graduate mentors, contributing to advanced NLP projects.- Lead author in research paper- Lead the development of AAVENUE, an NLP benchmark designed to detect biases in LLMs, particularly in African American Vernacular English- Published research papers, including those accepted at high-ranking AI conferences like the NeurIPS HS Conference 2024 (6% acceptance rate) and EMNLP 2024 in the NLP4PI workshop- Research has been cited by researchers at Microsoft, University of Washington, NYU, and University of Oxford

Conducting research at the University of Michigans SLED Lab alongside a PhD. Focusing on the development of web agents.

- AI-powered tool that summarizes complex academic papers into clear, concise summaries - Stock tool allowing users to input a stock ticker and instantly access relevant financial data - Backed by Microsoft with funding and support - Over 2.7k active users and 12k+ views - Used in 3+ classes and 3 clubs

- Created instructional materials for AP courses, focusing on strategies for achieving high scores.- Ensured all content complied with copyright regulations, maintaining legal and ethical standards.- Conducted quality testing of AI Tutor and Grader Chatbots, identifying issues and improving functionality.- Executed editing to enhance accuracy, consistency, and overall performance of AI-driven educational tools.