OpenSlate (acquired by DoubleVerify) provides a service to advertisers that allows them to control where their ads run on YouTube and Facebook. During my time there I built an ML pipeline in PySpark that allows OpenSlate to categorize text content at scale. Metadata from are large percentage of viewed YouTube videos are categorized on a daily basis with the models built and deployed with this framework.

• Worked with the ADP DataCloud team to enhance ADP’s products by integrating external data sources with ADP’s data. • Built Entity Resolution pipeline (People, Places and Organizations) to augment ADP's data.• Supported multiple dataset acquisitions. Estimated the value external datasets would add to ADP's dataset using sampling methodologies.• Worked with the New York Private Equity sales team to build custom dashboards for firms.

Built a dashboard that allows Sierra Club management to track how each regional chapter is performing on over 100 measures and allocate funds accordingly.

In my second year at Stony Brook I began freelancing as a Data Scientist through UpWork. I gained valuable experience during this time and completed the following projects for different clients:• Statistics project for The University of Massachusetts - Boston• Predictive hiring model built using Azure ML for Resource Pro• R Shiny analytics dashboard for The Sierra Club• R Shiny analytics dashboard for Applied Biomathematics• R Shiny analytics dashboard for the Faculty Student Association at Stony Brook

Working with Dr. Lev Ginzburg, Dr. Scott Ferson, and Dr. Nick Friedenberg of Applied Biomathematics was a formative experience in my career. During my time working with them I completed several projects:• Ecological Dashboard: Matthew Spencer, a professor of Ecology at the University of Liverpool recently attained a grant to build a software implementation of his ecological research on biodiversity. In collaboration with Applied Biomathematics Inc., I (working under Vivaldi Analytics) developed both a web application and R statistical package that calculate Spencer's measure of biodiversity for species abundance datasets. Complete with interactive charts and support for 3 types of data (scalar, interval-valued, and Monte Carlo iterates). The source code for this package can be found here: https://github.com/vivaldianalytics/shapechange• Interval Statistics calculator (IntStats): Dr. Scott Ferson's work focused on interval statistics - how to calculate traditional statistical measures such as expectation and variance on data with uncertain values. I created a calculator that generates visualizations and calculations on interval data using the formulas and algorithms that Scott and others had developed.• Theoretical Ecology research: I assisted Dr. Lev Ginzburg with his research into how an organism's surface area correlates with its metabolic rate.