Predictive Modeling: Developed an end-to-end Python platform using Scikit-learn, deploying a Random Forest Regressor for performance forecasting and a Gradient Boosting Classifier (with SMOTE) to predict promotion readiness. o Explainable AI (XAI) & Diagnostics: Implemented SHAP values with Logistic Regression to uncover core behavioral drivers of employee attrition, and applied Linear Regression to detect pay equity disparities. o Prescriptive Analytics: Built an automated recommendation engine that translates complex predictive outputs into actionable HR policies, enabling data-driven retention and talent development strategies.

Developed a Power BI dashboard to transform complex logistics data into operational insights, designing an optimized Star Schema and authoring complex DAX for performance correlations.

he objective was to build a Power BI dashboard that provides: Clear KPI tracking Risk segmentation Demographic and behavioral analysis Interactive dynamic reporting 📊 KPI Metrics The dashboard includes the following key performance indicators: Total Policies – Measures the size of the customer base Total Claim Amount – Overall financial exposure Claim Frequency – Indicates how often claims occur Average Claim Amount – Measures claim severity Gender-wise Policy Distribution – Customer segmentation

The goal to answer: Where is the revenue coming from How efficient is the cash collection? Where are the financial risks

This project demonstrates the design and implementation of a SQL-based Data Warehouse using Swiggy order data. The goal was to transform raw transactional data into a structured Star Schema and extract business insights using analytical SQL queries.