Imperial CS student interested in solving real-world challenges with ML and exploring quantitative finance.
Built multi-agent AI workflows with feedback loops for a partner integration platform. Improved system accuracy from 70% to 95%+ and reduced onboarding from 6 months to <5 days.
Full-stack web app using FinBERT for NLP sentiment analysis, scraping financial news, and displaying visual charts for stock market insights.
Developed a portfolio allocation algorithm balancing profit and volatility for the QRT challenge. Used linear regression and gradient ascent to optimize allocation parameters.
Platform for pathologists to analyze cancer risk images with CNN-generated heat maps, highlighting high-risk areas for early detection.