Advanced Data Warehouse Project - Sales & Returns Analysis

Built an end-to-end data pipeline integrating OLTP and Data Warehouse systems across MySQL and Oracle, transforming 51K+ transactional records into structured analytical insights for reporting and decision-making.

• Designed and implemented OLTP (normalized) and Data Warehouse (STAR schema) systems across two database platforms
• Built ETL pipeline to extract, transform, and load data from OLTP to DW environment
• Implemented Change Data Capture (CDC) for incremental loading, reducing processing time by ~30%
• Performed complex SQL transformations including joins, aggregations, and schema mapping
• Designed fact and dimension tables (customer, product, date, location, ship mode) for analytical queries
• Developed Tableau dashboards tracking 8–10 key KPIs across sales, profit, and returns

IBM Capstone Project (Developer Survey Analysis)

Analyzed real-world developer survey datasets to uncover technology trends and generate actionable insights. Focused on data cleaning, exploratory analysis, and dashboard development for KPI tracking.

• Cleaned and transformed large-scale survey datasets using Python (Pandas) to prepare structured data for analysis.
• Performed exploratory data analysis (EDA) to identify trends in programming languages, cloud platforms, and database technologies.
• Queried and aggregated data using SQL to extract key metrics for reporting and analysis.
• Built interactive dashboards in IBM Cognos Analytics to track KPIs and visualize technology adoption trends.
• Translated analytical findings into actionable insights to support data-driven decision-making.

View on GitHub ↗

Bank Customer Churn Prediction Project

Tools:Python (Pandas, Scikit-learn, Matplotlib, Seaborn), Power BI, Jupyter Notebook

• Developed a complete end-to-end customer churn prediction system using machine learning and visual analytics.
• Built a classification model using Logistic Regression, Decision Tree, and Random Forest to predict whether a customer will churn based on their demographics, account activity, and balance.
• Achieved 87% accuracy and a significant improvement in recall and F1-score using Random Forest after preprocessing and feature scaling.
• Performed EDA and feature engineering, including missing value checks, categorical encoding, and feature scaling.
• Designed a Power BI dashboard to visualize churn distribution across gender, age, credit score groups, and tenure — revealing key business insights.
• Evaluated all models using Accuracy, Precision, Recall, F1-Score, and Confusion Matrix to determine the best-performing algorithm.

View on GitHub ↗

NYC Turnstile Data Analysis

Tools:Apache Spark, PySpark, Kafka, SparkML, MongoDB, Python, SQL, Pandas, Seaborn, Matplotlib

• Analyzed over 13 million records of NYC MTA Subway turnstile data to uncover ridership trends and station-level traffic patterns using PySpark and Pandas.
• Built an end-to-end real-time data pipeline using Apache Kafka, PySpark Structured Streaming, and MongoDB for live prediction of subway foot traffic.
• Developed and deployed machine learning models (Random Forest, Decision Tree, Linear Regression) using SparkML to predict hourly passenger entries/exits with 93.36% accuracy.
• Engineered custom Kafka producer scripts to simulate realistic subway traffic, publishing streaming data every 3–5 seconds.
• Designed insightful visualizations including heatmaps, boxplots, and bar charts to reveal peak hours, seasonal usage, and holiday traffic dips using Seaborn and Matplotlib.
• Implemented delta-based transformation logic to convert cumulative turnstile counts into accurate interval-based foot traffic data.
• Managed structured and semi-structured data using MongoDB to enable flexible querying and integration with Spark SQL.

View on GitHub ↗

NICE Cruise Case Study

Tools: MySQL, PHP, HTML, CSS, JS, XAMPP

• Designed and Implemented robust relational database schema to manage cruise operations.
• Developed a RESTful web application using PHP, MySQL, HTML, CSS and JS.
• Enhanced security by implementing SQL injection prevention, password hashing, XSS protection and secure session handling.
• Integrated CRUD operations for user and admin interactions supporting booking, updating and cancelling functionalities.
• Created SQL queries providing insights into passenger behavior, trip occupancy and revenue trends.
• Built a data visualization module to analyze passenger statistics through interactive graphs and charts.