• Founded a market R&D organization that tracks, organizes, and catalogues ~300 startups in the responsible AI enablement space.• Consulted for various international governments, venture capital firms, and startups looking to understand the dynamics of this ecosystem.• Gave interviews and conference talks, led panels to raise awareness and promote companies in the database.

• Led product development for Vanguard's premier Generative AI product (fixed income investor-centric RAG) from scratch with a full, open-source LLM stack. Project cited by CIO in various conferences.• Designed and built a system of LLMs featuring a novel, research-driven approach ("Mixture of Agents (MoA)," later researched by Together AI) that analyzes and produces insight from 30,000+ documents in under 60 seconds.• Product features a novel, research-driven approach for interacting with combinations of LLMs (called Mixture of Agents, paper pending).• Resultant product outperforms most modern equivalents and currently has ~100 users across use cases in active/passive funds and financial advisory services.

• Leveraged Home Data Mortgage Act (HMDA) data to perform extensive analyses on the state of mortgage fairness in the USA (including intersectional and geographic fairness).• Implemented clustering techniques and designed metrics in a first-of-its-kind analysis of lender marketing fairness.

• Writer for "Towards Data Science" and "Towards AI," two popular online publications about topics in data science, machine learning, and artificial intelligence.

• Implemented techniques in time series analysis including ARIMA models and online learning to perform statistical backtests and proof-of-concepts for several investment strategies.• Built a Python API to more efficiently interface with Refinitiv's financial data and streamline several common use cases.

• Implemented a Random Forest model to identify the best building targets for Starry's marketing team.• Utilized data stored in Snowflake, built models in R.

• Lead team of undergraduate students to identify cause of churn in Teach for America's (TFA) candidate matching procedure.• Performed clustering analyses (KMeans, DBSCAN, etc.) to identify application groupings.• Built a strong Random Forest model that exceeded 95% (0.94 AUC) on out-of-sample data.• Designed validation methodology and visualizations to pinpoint root cause and characterize TFA's locus of control.

• Covered 150+ healthcare tickers for COVID-19 related financial updates• Performed in-depth industry research on the video gaming, chiropractic, and home furnishing industries• Implemented techniques in data science on third-party app data (AppAnnie, Sensor Tower)

• Worked with Python and Amazon Web Services on IHS Markit’s loan reconciliation software• Collaborated with Market Reconciliation Services (MRS) to automate loan tracking processes for 10+ financial clients and cut 50+ hours of manual labor• Finished service will add over $3M to WSO revenue

• Performed in-depth market research and feature analysis in voice technology• Created modeling algorithms to classify patients by a set of variables relating to patient activation• Developed Amazon Alexa Skills backend and API connectivity in NodeJS