• Selected to intern for 2080 Ventures, a VC accelerator focused on providing growth capital ($5M – $15M) to companies operating in the emerging markets across Asia and Eastern Europe to provide startup acceleration aimed at the local markets• Support 2080 in company outreach, research and evaluating potential investments on qualitative and quantitative factors• Develop deep learning algorithms for startup evaluation

• Elected to be the First Year Representative of Will Rice College.• Lead a 10-person committee to organize events and facilitate connections

• Continuing my PPI complex prediction and categorization project with Dr. Hamid Teimouri and Professor Anatoly Kolomeisky• Exploring the correlation between protein frustration and PPI formation using machine learning algorithms

• Studied and researched ML survival analysis models, including DeepSurv, the Rolling Extrapolation System, etc.• Implemented a 3-stage regression ML model to quantify various hospitals' patient population difficulty that achieved 85 percent accuracy• Co-authored the Unannounced Inspection GPT and won the 2024 National AI Healthcare Competition

• Assisted Chiara Smorada, Matthew Sima, Shan Huang, and Professor Peter Jaffé (Civil and Environmental Engineering) with their research projects regarding the A6 bacteria group’s ability to degrade PFAS compounds• Performed DNA extraction, PFAS analysis, qPCR, LCMS, NMR, and iron analysis on 4 incubation experiments• Presented research findings at lab meetings and submitted the final report to the Princeton Laboratory Learning Program• Presented research findings at BRICS 2023 as a Jaffé lab intern• Attended the ICON8 conference at Princeton University as a poster presenter

• Collaborated with with the robotics department for developing medical robots• Did extensive research on past detection networks–hourglass CNN and ISEM models• Improved the speed of face detection by 4.8 times by optimizing the shape of the convolutional network• Reached out to and collaborated with experts from Beijing University of Traditional Medicine to enrich our dataset• Enhanced detection accuracy from 82% to 98.9% by implementing the yolov5l model for face acupuncture point detection