I turn messy data into decisions that save companies money and grow revenue.
- 🚗 Analyzed 195K+ Uber rides — uncovered an 81% rider drop-off by Day 30 through SQL cohort retention analysis
- ⛽ Built a fraud detection pipeline across 1,00,000 transactions — uncovered a 27.5% fraud rate and physically proved 16,200L of under-dispensed fuel using sensor data
- 📈 Forecasted $192K Q1 revenue using ARIMA time-series modeling
- 👥 Segmented riders using K-Means clustering — isolated a High Value tier averaging 4x the platform's average fare
- 🎓 Microsoft Power BI Data Analyst Professional Certificate (verify) · Google Data Analytics Professional Certificate (verify)
🎯 Open to entry-level Data Analyst / MIS Analyst roles — Remote / Hybrid — available from August 2026
Built an end-to-end fraud detection pipeline analyzing 1,00,000 transactions across 50 fuel stations in 3 cities — uncovered 27.5% fraud rate (peaking at 51.2% in Delhi), physically proved 16,200L of under-dispensed fuel via sensor data, and validated 3 business hypotheses linking fraud to customer retention (fraud-affected customers return at 18% vs. 36% for clean transactions).
Analyzed 195K+ NYC Uber rides to diagnose retention, cancellation, and pricing patterns. Wrote SQL window-function queries for cohort retention and completion-funnel analysis — uncovering an 81% rider drop-off by Day 30 (34% D7 → 19% D30 retention). Applied KMeans clustering to segment riders into Low/Mid/High value tiers, isolating a High Value segment averaging nearly 4x the platform's average fare. Built a 3-page Power BI dashboard with 5 DAX measures for executive-level reporting.
ML pipeline on 7,043 customers predicting 26.54% churn — trained 4 models with SMOTE balancing; Random Forest selected as optimal.
End-to-end fintech analytics project on Buy Now Pay Later lending data — cleaned and analyzed loan-level data in Python, then built a Power BI dashboard segmenting customers into High/Medium/Low credit risk tiers, tracking default rate, and surfacing which merchant categories and providers drive the highest loan volume and risk exposure.
👉
Programming & Databases:
Libraries & Analytics:
BI & Reporting:
Tools:
💬 Got a role, a project, or just want to talk data? Reach out to me through LinkedIn or Email — always happy to connect.