Technical Highlights

• •Built end-to-end ETL pipeline integrating 5+ public health datasets (CDC PLACES, County Health Rankings, HRSA, and USDA) into Snowflake using dbt with staging, intermediate, and mart layers.
• •Engineered 10+ derived variables across dbt staging, intermediate, and mart layers; performed data quality validation checks via dbt generic tests on staging models.
• •Developed interactive Plotly Dash dashboard visualizing shortage severity, disease burden, and preventable hospitalizations, including a KPI summary row with a custom dark theme.
• •Identified poverty rate (r=0.33) and high disease burden counties (average 482 excess stays above the national average) as strongest predictors of preventable hospitalizations.
• •Conducted exploratory data analysis and inferential statistical testing in Python (summary statistics, correlations, t-tests, ANOVA, chi-squared), and SQL analyses to answer 5 research questions across 2,957 US counties.