Unlocking Insights in Healthcare

Engineered a healthcare decision-making system by building a predictive analytics system that achieved 97% accuracy in patient risk assessment and uncovered $20,000+ cost-saving opportunities through advanced clustering and ML modeling. Transformed raw patient data into an interactive dashboard for real-time monitoring of critical healthcare metrics.

Motor Vehicle Collisions

Leveraged advanced geospatial analysis and statistical testing to identify critical crash hotspots and high-risk patterns in NYC's motor vehicle collisions. Uncovered statistically significant correlations between accident severity, location, and timing (p < 0.001), enabling data-driven recommendations for emergency response optimization and targeted safety interventions.

Power BI Showroom

Storytelling through Power BI dashboards. Transformed raw data into interactive visualizations across various industries.

Competitive Intelligence Analysis for Kimball International

Drove data-backed market strategy for Kimball International by analyzing competitor pricing, features, and customer behavior across digital channels. Identified untapped $500-700 mid-tier market opportunity and revealed 47% surge in digital-first purchases, enabling strategic positioning in the growing ergonomic office chair segment.

Deep Marketing Data Analysis

Optimized bank marketing campaign ROI by uncovering that 78% of clients with >375-second interactions opened term deposits, versus 32% below threshold. Built ML models reaching 85% accuracy in predicting client conversion, enabling targeted outreach that increased successful deposits.

ETL Pipeline: front & backend

Built end-to-end automated ETL pipeline integrating Alpha Vantage API with BigQuery for real-time stock analysis. Orchestrated daily data workflows using Apache Airflow and created dynamic Power BI dashboards featuring 12+ key trading metrics, enabling instant market insights.

Topological Data Analysis

Coded application of Topological Data Analysis (TDA) for detecting market crash patterns in S&P500 data. Developed Python algorithm using persistent homology and Wasserstein distance to identify structural similarities between historical market periods.

This Website

Built this sleek portfolio from scratch using Astro, TypeScript, CSS and HTML. Clean design, fast load times, and fully open source - grab it from my GitHub if you want your own minimalist showcase!