Projects
Here are a few of the data projects I’ve worked on:
Citi Bike Strategy Dashboard
Analyzed NYC bike usage and weather trends to identify availability gaps and optimize bike distribution across stations. Used geospatial mapping to highlight underperforming locations and propose operational improvements.
Tools Used: Python, Pandas, Plotly, Kepler.gl, Streamlit
World Happiness Report Analysis
Analyzed five years of global happiness data (2015–2019) to uncover which economic, health, and social factors most influence happiness levels worldwide. Used Python for wrangling, clustering, and correlation analysis; Tableau for geospatial dashboards and insight delivery.
Tools Used: Python, Pandas, Jupyter Notebook, Tableau, Excel
Instacart User Behavior Analysis
Analyzed 32M+ grocery orders to uncover trends in customer behavior and guide targeted marketing. Identified purchase patterns by day, hour, price sensitivity, income, and marital status. Delivered strategic recommendations to improve segmentation and timing of campaigns.
Tools Used: Python, Pandas, NumPy, Seaborn, Jupyter Notebook, Excel
Influenza Season Staffing Plan
Analyzed influenza mortality trends across U.S. states using CDC and Census data to forecast seasonal staffing needs for hospitals and clinics. Built visualizations in Tableau to identify high-risk regions and inform medical staff distribution during peak flu months.
Tools Used: Excel, Tableau
Rockbuster Stealth LLC – SQL Case Study
Used SQL to help a classic video rental company transition to a digital model by analyzing rental patterns, top-performing films, and high-value customers. Delivered strategic insights using PostgreSQL queries, a custom data dictionary, and a full business presentation.
Tools Used: PostgreSQL, SQL Joins, CTEs, Subqueries, PDF Reports, PowerPoint
View SQL Workbook (Google Sheets)
Summit Bank Customer Retention Analysis
In this solo Excel project, I explored client data from Summit Bank to uncover patterns in customer churn. After cleaning and analyzing 991 customer records, I identified the top factors influencing client exit and built a decision tree to visualize risk levels.
- Conducted data quality checks and cleaned client-level attributes
- Segmented customers by churn status to compare behavioral patterns
- Identified churn predictors: inactivity, short tenure, poor credit, and high balance
- Created a decision tree to highlight top churn risk profiles