Research

Cognibot

Jan 2025-Present

• Developed a React + Vite + TailwindCSS web application enabling dementia patients and caregivers to visualize conversational history, biomarker metrics, and care progress, improving accessibility and user satisfaction scores in pilot tests of 8 users.
• Built and secured a Python/Django REST Framework backend with PostgreSQL, supporting user authentication, data management, and API scalability for multi-user healthcare environments.
• Reduced end-to-end latency of an LLM conversational chatbot by ~500ms by developing and implementing Google Gemini API with custom audio streaming, enhancing conversational responsiveness and patient usability.

ClassTranscribe Accessible Technology in Education Development

Jan 2023-Dec 2023

• Improved accessibility options of the web application ClassTranscribe using tools such as React and JavaScript to facilitate greater accessibility options for all users.

Assessing Risk Factors For Chronic Fatigue Syndrome From Long COVID

June 2022-Aug 2022

• Designed and programmed programmed machine learning/AI algorithms and methods in R to analyze over 300 study responses to predict risk factors for the disease chronic fatigue syndrome.
• Utilized iterative learning methods to test and develop the model, eventually reaching over 90% accuracies.
• Authored new and original research paper on findings, published in the journal Psych.

Predicting Chronic Fatigue Syndrome Using Cytokine Correlations

June 2022-Aug 2022

• Developed statistical models in R based on correlation of medical data to predict development of CFS.
• Conducted dozens of experiments to assess the efficacy of the created models.
• Enhanced and extended a research paper on the model, published in an IEEE conference.