- Present MBR and other high-impact monthly reports to internal and external stakeholders, providing actionable insights, improving cross-team decision-making and leading to more targeted fraud prevention efforts and improving KPIs. - Took full ownership of fraud strategy creation in July, driving significant increases in detection rates, cutting fraud losses and halting major months-ongoing trends in Outside Card Not Present and In-Store Card Not Present areas while balancing customer experience. - Increase fraud detection for In-Store Card Not Present by 60% (Full Line Stores), 450% (Rack Stores), and 150% (online) since July. - Create decline rules for In-Store Card Present fraud, boosting detection rates by 40% since August. - Created decline rules for Outside Card Not Present with No Chargeback, increasing fraud detection by 57% (US) and 38% (intl) since August. - Conduct exploratory data analysis on all credit card transactions in all areas of credit card services, discovering gaps in existing strategies. - Perform Feature Engineering and Feature Selection with Random Forest Models in Python to select features, create decline rules, reducing monthly online fraud losses and increasing detection. - Visualize data story in Tableau dashboards for executives and fraud operations team, and informed strategy decisions that reduced fraud losses while maintaining a low impact on good customers. - Collaborate with Fraud Operations team to discover recent trends or events in order to swiftly create all ad-hoc strategy.

- Analyzed Ski Resort's data to create Random Forest Model in Python to target important facilities that could command a higher ticket price but appropriate relative to its competitors. Determined which facilities to add that would support a 350k revenue increase over the next year.- Performed Exploratory Data Analysis and created logistic regression model to classify heart-disease detection using using patient health data. Overall accuracy was 81%, 20% of the time predicting a false positive, while the false negatives was about 18%.- Created XGBoost Machine Learning Model to predict a National Grocery Store's sales in item units by item given historical store data, item data, promotional data, holiday details, and oil prices by date with a Training RMSE of 0.147 and a Test RMSE is 0.185 in item units.- Created Spark ML Pipelines on DataBricks, fitting a logistic regression and a random forest.- Created Gini Impurity Model and Random Forest Model to predict percent of customers that would purchase new line of Coffee by RR Diner and presented insights to inform decision to introduce the new product.- Analyzed London housing data with 45 variables to create linear regression model to predict housing prices.

- Led product analytics initiatives, driving a 20% lift in product utilization by developing interactive Tableau dashboards to track feature performance and adoption.- Created 90 day agent ramp up analysis to show value of XSELL in improving Verizon agent ramp to proficiency which was then implemented as a standard analysis across the company for client reporting.- Conducted A/B testing analysis on new product features, increasing feature engagement by 15%, which contributed to overall customer satisfaction improvements. - Automated SQL processes for client reports, reducing time spent on weekly analytics, improving operational efficiency by 90%. - Created Top Perform Analysis for CFO of XSELL to present to Verizon Executives to show how XSELL usage correlates with top performing agents.- Created Quarterly Business Review for Verizon Executives to show value of HiPer and maintain faith in the product.- Partnered with executives to develop business case analyses, securing long-term contracts worth $2M by demonstrating product ROI through data-driven insights.

- Performed Exploratory Analysis of American Energy Market Regulator Data Sets using SQL to describe, compare and visualize the reliability of 18 electric companies over 2 years in terms of number of outages, and energy lost.- Used Excel and Python to create a best fit regression model to predict untimely asset failure which would prevent unexpected asset failure by around 60%, visualized findings with Excel and Python, and provided recommendations for managing significant variables.- Developed short-term plan for ChemCorp, in response to increased competition, to maintain market share, recoup 10% of lost sales revenue, and retain customer base by the end of 2020 through leveraging customer strategy and product divestment. Visualized findings in Power BI dashboard.

- Analyzed and cleaned data from ~180 alumni records using Excel, identifying key variables to target high-potential leads.- Conducted data-driven lead prioritization using CRM data and external datasets, improving targeting efficiency.- Utilized tools like SalesGenie to segment small businesses based on performance metrics and demographic indicators.- Automated the review of business documents (e.g., licenses, P&L statements, tax returns) to streamline applicant evaluation.- Gained hands-on experience in data management, customer segmentation, and insight generation to support decision-making processes.