Cambridge, Massachusetts, United States
• We unify unstructured internal data and pair with property, mortgage, and client life intel to provide a supercharged, living database that drives revenue for real estate brokerages. • Backed by NAR REACH • Harvard Law Entrepreneurship Competition Finalist: Gained access to mentorship from 8 Harvard Law School attorneys. • MIT $100K Finalist • MIT Sandbox Grant Recipient
Led dataset exploration, feature analysis, and multi-model experimentation (LLM agents, GANs, diffusion models) to create synthetic malware behavior data for bank-fraud defense research
Developed a full-stack application (TypeScript, Java, React) for a cloud-based cross-domain service, significantly improving the speed and simplicity of customers' schema management for data transfer. I also integrated existing APIs in a way that abstracted away previous complexity for customers (Java, Go)
Used deep learning to establish a reinforcement learning based control system paired with a novel safety filter, improving the accuracy, adaptability, and risk management of microgrids.
Built a state-of-the-art text-to-SQL language model to access data within a database, generating visualizations and analytical insights from natural language inputs.