Project Portfolio

  • A/B Testing Landing Page Engagement for LunarTech.ai

    Download Notebook

    View Notebook on Github

    This project involved designing and executing an A/B test to evaluate the impact of different call-to-action (CTA) texts on click-through rates (CTR) for LunarTech.ai. I outlined the experiment design and hypothesis, conducted a power analysis, and analyzed the A/B test results. I provided recommendations based on both statistical and practical business significance of the test results.

    This project demonstrates a methodical approach to data-driven decision-making for optimizing marketing strategies.

    Skills & Tools

    A/B Testing | Statistical Analysis | Hypothesis Testing | Significance Testing | Data Visualization | Research Design

    Completed: January 2025

  • Building Classification Models to Predict Clients Who are Likely to Default on a Loan and Giving Bank Recommendations on Feature Importance in Approval Process

    Download the PowerPoint Report

    Built, trained, tested, and optimized seven different classification models to predict likelihood of loan default. Utilized EDA, Decision Tree, Random Forest, and Logistic Regression to determine feature importance and prevent monetary loss to the bank.

    Chose the optimal model based on performance and different measures of accuracy. Presented findings, proposed solution design, and risk/cost evaluation to panel of Principal Data Scientists and business stakeholders. Recommended model be deployed to originations team via API.

    Skills & Tools

    Logistic Regression | Decision Trees | Random Forest | Bagging | Hypothesis Testing

    Completed: January 2022

  • Building a Recommendation System to Recommend the Best Amazon Products to Users

    This project involved recommending the best Amazon products available to users based on past rating data using recommendation systems techniques.

    Skills & Tools

    Rank Based Recommendation Systems | Similarity Based Recommendation Systems | Matrix Factorization Based Recommendation Systems

    Completed: December 2021

  • Recognizing House Number Digits from Street View Images Using Neural Networks

    This project involved recognizing street view housing number digits using artificial and convolutional neural networks.

    Skills & Tools

    Artificial Neural Networks | Convolutional Neural Networks

    Completed: December 2021

  • Predicting Lead to Customer Conversion

    This project involved identifying which leads were more likely to convert to paid customers based on attributes of leads and their interaction details.

    Skills & Tools

    Decision Trees | Random Forest

    Completed: December 2021

  • Predicting House Prices in Boston and Predicting Hotel Bookings Cancellations

    This project involved predicting house prices in the Boston metropolitan area based on the features of the property and its locality using regression techniques.

    Skills & Tools

    Linear Regression | Logistic Regression | kNN

    Completed: November 2021

  • Applying Dimensionality Reduction on the Auto-mpg Dataset and Segmenting Bank Customers

    Download the Project Notebook

    The first part of this project involved exploring the Auto-mpg dataset by applying dimensionality reduction techniques and visualizing the data in lower dimensions to extract insights.

    The second part involved segmenting a bank's customers to help the bank upgrade the service delivery model and ensure that customers' queries were resolved faster.

    Skills & Tools

    PCA | t-SNE | Clustering

    Completed: November 2021

  • Analyzing Marketing Campaigns to Help the CMO Devise the Next Best Marketing Strategy

    This project aimed to analyze marketing campaign and sales data to address important business questions related to customer demographics, product preferences, channel performances, etc., and help the Chief Marketing Officer in devising the next best marketing strategy.

    Skills & Tools

    Exploratory Data Analysis | Data Visualization | Statistics

    Completed: November 2021

  • MPA Capstone Project with the RAND Corporation

    Examined how the US Intelligence Community can operate in an unclassified environment through remote work. Identified the technological, legal, policy, financial, and security considerations that would allow employees to engage in missions outside of classified environments.

    Analyzed Intelligence Community risks from insider threats and external adversaries using cyber collection techniques, and how remote work would impact this. Interviewed telework management officials from FEMA, GSA, IRS, NRC, and NSF regarding best practices, lessons learned, investment costs, ROI, and how telework was implemented.

    Condensed a research report on findings and disseminated to the RAND Corporation. Presented findings to high level officials at the RAND Corporation for later distribution to their Intelligence Community client.

    Skills & Tools

    Data Gathering | Data Analysis | Data Visualization | Statistics | Research Design

    Completed: July 2016