I lead the Data Science chapter at Syrup where we leverage AI and ML to build the next generation of Inventory Optimization software

I led a full stack production team of Data Engineers, Data Scientists, ML Engineers, and Full-Stack Developers responsible for developing, deploying, maintaining, and enhancing the models and pipelines powering our Enterprise Software-as-a-Service and Data Products.

I managed and led a team of data scientists at BlackLocus, The Home Depot Innovation Lab. We worked with Big Data and leveraged cloud computing platforms to build, deliver, and scale ML models that solved complex business problems related to pricing, competitive intelligence, and product matching for The Home Depot. In my role, I was responsible for leading all aspects of the model lifecycle from ideation to deployment, driving collaboration with cross-functional partners (Engineering and Product teams). I developed strategic initiatives aimed at unlocking short and long run value of Data Science work to the enterprise. I advocated for adoption and scaling of solutions with key business stakeholders, and set/report on Data Science driven OKRs to senior leadership.

I led a team of quantitative risk analysts/data scientists tasked with validating complex Statistical/ML models used in predicting credit/market risk, assessing Value-at-risk/capital planning, and evaluating stress testing scenarios at the enterprise level. The models in my portfolio were used for crucial business processes with high visibility to the C-suite and Board, including simulating credit/market losses, stress testing, enterprise liquidity testing, and capital planning.In my role as Director, I was responsible for managing the entire model validation life cycle, including issuing validation and model annual review reports, ongoing performance monitoring, and governance tasks. I provided technical leadership to my team in understanding and replicating the model development and implementation processes, often leading to model improvements that reduced risk to the Enterprise.My responsibilities also included managing day-to-day relationships with our model owners (including Product and Engineering teams) and external consultants, interfacing with regulatory control partners, and leading training and career development efforts for staff.

Worked with company data to replicate, test, and enhance various statistical and ML models to help manage enterprise model risk. As part of this, I managed small teams of 2-3 analysis throughout the project lifecycle, and led authorship of model validation reports.

In this role, I built and implemented statistical models for merger analysis and damages calculations, which enabled clients to clear mergers and avoid multi-billion dollar fees. In developing these models, I queried and collected data from the clients, performed data cleaning and aggregation, and conducted Exploratory Data Analysis/Data Visualization to identify important relationships in the data. I presented all model results to management, clients, expert partners, and regulators.I also managed high-performing teams of 2-5 Economists, completing tasks on time and within budget for projects with budgets ranging from $100,000 to $5 million.

As a Data Scientist/Consultant, I worked on various engagements for public utilities clients. Example projects include: wrangling big datasets (high frequency reading from smart meters) for use in predictive modeling, which resulted in 10%-15% reductions in energy usage during peak demand periods; designing surveys, sampling methodology, and conducting statistical analysis for energy efficiency/demand response projects.I also contributed to business development activities, including responding to RFPs and developing project budget plans, resulting in $1 million+ of new business.

I served as a Data Scientist/Consultant on multiple engagements for government and law firm clients. My role was to serve as a technical expert providing statistical/ML model solutions for clients using techniques such as logistic and multinomial logistic analysis, factor analysis and survival analysis. I also was responsible for conducting program evaluations using econometric techniques such as instrumental variables, GMM, Regression Discontinuity Design, and Difference-in-Difference modeling.

I taught two sections of Statistics (35 students each) on a tri-weekly basis. Assigned and graded problem sets and exams. Regularly met with students to discuss progress and provide extra assistance.

I was a summer research fellow responsible for analyzing the influence of housing market decline on individual attitudes towards home ownership. My responsibilities included compiling and cleaning consumer survey data, Census data, and housing loans and house price indices data. I then developed econometric models and presented the results to the entire Boston Fed Research Department.

I developed econometric methods for casual inference in fully endogenous structural systems. To test these methods, I coded equations and conducted simulation analysis using MATLAB. I then applied the techniques to estimate returns-to-schooling using data from the National Longitudinal Survey of Youth (NLSY).

I was a research assistant responsible for an investigation of the impact of cultural beliefs on employment decisions. As part of this work, I assembled and cleaned General Social Survey (GSS) and Bureau of Labor Statistics (BLS) datasets and experimented with different econometric models to select the best performing model.

I spent a summer conducting independent research on housing economics and racial dynamics in Kingston, NY. As part of this work, I consolidated data on local area housing prices and Census data on neighborhood demographics, and built spatial regression models to predict the relationships between housing prices and demographics. The results were published in the co-authored paper “Racial Preferences in a Small Urban Housing Market,” published Housing Policy Debate, Vol. 22 Issue 2 pp. 241-269