Instructor for CSCI 6443: Data Mining (graduate level). Responsible for developing curriculum and teaching this course that is focused on key data mining techniques, including classification, clustering, association rules, graph analysis, and text mining.

I provide strategic advice and guidance to founders and operators within the Primary portfolio and broader network. Primary is New York City’s premier early-stage venture firm working alongside founders to build unicorns like Alloy, Alma, Chief, Dandy Electric, K Health, Slice, and many more.

Player-coach for a team of 7 data scientists at Enigma. Key contributor to closing deals with prospects during customer evaluations. Shored up Enigma's MLOps practices by launching the company’s first model validation and monitoring processes.

I led an organization of 20 data scientists and analysts responsible for machine learning, data, and all fraud decisioning logic. Led the design and implementation of Vesta’s first cloud-based fraud decisioning engine using Databricks, Spark Structured Streaming, MLFlow, and Azure.

I led the Fraud Data Science team within Capital One’s Credit Card line of business. Some highlights:* The development, monitoring and updating of over a dozen machine learning models for detecting all fraud types that are worth over $250M in total annual value* The development of an interactive graph analysis software being used by over 150 Fraud and Anti-Money Laundering Investigators to uncover fraud rings and other high risk behavior.

I led the internal data science consulting team within Capital One Labs, building ML and data products for a wide variety of use cases, including fraud, anti-money laundering, auto finance, customer experience, and commercial lending. Some highlights:* A partnership with our Anti-Money Laundering team has dramatically improved our ability to detect activity related to human trafficking, changed the way the AML organization develops new defenses.* A partnership between Capital One’s Labs and Auto Finance data scientists to re-implement the valuations modeling framework that used for credit decisioning across the entire auto lending portfolio.