An explanation-first HR analytics system that reconstructs why employee exit becomes rational. Instead of predicting attrition, it generates human-readable exit narratives by decomposing pressure and retention forces, adding peer context and counterfactual interventions to reveal how stability erodes over time.

RFM-based customer segmentation analysis for an e-commerce dataset. Includes data cleaning, exploratory analysis, Recency-Frequency-Monetary scoring, segment classification, visual dashboards, and strategic business insights. Designed to identify high-value customers and guide targeted marketing actions

A Python package for calculating and visualizing cohort retention and other cohort-based metrics with interactive Plotly charts.

SQL-based business analytics on the Online Retail II dataset, answering core customer, revenue, retention, and product performance questions.

Full Product Data Science Workflow: Case study using K-Means Clustering to isolate a critical churn bottleneck (L4, 75% fail rate) in a mobile game's early player journey. Features LLM integration (AI Playtester) to generate a final, validated A/B test plan (the +2 Moves fix) targeting immediate retention uplift.

A complete Streamlit + Machine Learning + SHAP + NLP project to analyze, predict, and improve player retention in games. This project simulates a game environment, models churn behavior, and provides insights using SHAP, NLP word clouds, and strategy simulators.