Greater Richmond Region
I recently graduated from a M.S. in Computational Operations Research at William & Mary. I work as a data scientist in the credit industry.
-Extended the use of a custom data quality tool that sends twice-daily automated alerts to the click-based-outbound (CBO) business team on the health of the performance reports for the Discovery and Gmail advertising channels. The health of these performance reports is crucial for the CBO team, as they use these pipelines on a daily basis. -Used object-oriented programming to implement methods repeatedly for the various layers of the data quality tool. Worked closely with software engineers and other technology team members. -Presented the insights and value added of the custom data quality tool to senior leaders in technology and product. Programs used: Databricks, Snowflake, OneLake, AWS Languages used: Pyspark, SQL
-Designed a clustering methodology to pair similar targeted and non-targeted individuals and compare their conversion rates derived from a campaign for a drug. This was an exploratory project that would help Medicx develop a new product aimed at analyzing how the company positively impacts conversion rates via targeting relevant groups of people. -Extracted data to examine how backfill impacted campaign data. Used statistical methods to determine the ideal wait time to analyze campaign data, taking into account the impact of this backfill. Programs used: MySQL, Microsoft Excel
-In the modeling team, I designed Value at Risk models, forecasting the maximum amount of loss we would expect on a given day, making recommendations on cash reserves given our risk appetite. Used sensitivity analysis to tune this model to adjust for different assumptions, such as whether we want to focus on historical occurrences or whether we wanted to make normality assumptions. -I helped manage $25,000 from the UR’s endowment by investing in exchange-traded funds. Moreover, I oversaw each region-divided team, while also leading top-down analyses in the Latin America team. Programs used: R, Microsoft excel, Google sheets, Microsoft PowerPoint
-Determined the revenue contribution of repeat customers and identified how to target such groups. -Analyzed and synthesize approximately 60 million rows of data at a time. -Used probability trees to determine which customers were more likely to become repeat customers using a sample of 1.5 years of data. -Evaluated the profitability of a part of the Mexican business unit. -Identified and drafted a concrete expense reduction plan that will save the company $50,000 per year. Programs used: R, Microsoft Excel, Microsoft PowerPoint