My Projects

End-to-end retail analytics solution — interactive Power BI dashboards analyzing $45M+ in operational sales data across stores, districts, and product categories, with overview, store, district, and item analysis layers.

• Designed a scalable star-schema model — Sales fact with Store, Item, District, and Time dimensions.
• Built advanced DAX measures: Total Sales, Sales Variance %, Gross Margin, Sales per Sq Ft, YoY KPIs.
• Multi-level analysis identifying top/bottom performers and regional sales trends.
• What-If analysis to simulate pricing changes and forecast revenue impact.

Integrated multi-source datasets (FAOSTAT, World Bank WDI, emissions) across 180+ countries (1980–2023) to analyze food security, inequality, and supply stability at a global scale.

• Engineered Food Affordability Index and Nutrient Supply Score.
• Identified 3 global food crisis periods; methane-heavy economies showed 25–40% higher price volatility.
• Geospatial analysis classifying countries into risk tiers for policy insights.

Analyzed EV adoption, charging infrastructure, and PM2.5 pollution across 58 California counties — linking infrastructure investment to environmental outcomes.

• Unified EPA and ARB datasets into a single GeoDataFrame for large-scale geospatial analysis.
• Quantified up to 30% lower pollution levels in high EV-adoption regions.
• Built interactive Folium dashboards supporting regional planning decisions.

Exploratory data analysis and visualization of 398 mass shooting incidents across the U.S. — uncovering spatial, temporal, and statistical patterns to support evidence-based public safety discussion.

Combined Bitcoin historical price data with Twitter, Reddit, and news sentiment to test whether public sentiment improves price prediction — using walk-forward backtesting on ARIMA and LSTM models.

Capstone project for the IBM Data Science Professional Certificate — an end-to-end applied data science project covering data collection, wrangling, analysis, visualization, and machine learning.