My Portfolio
Waze Churn Prediction and Recommendation App
Developed an end-to-end machine learning pipeline to predict and prevent user churn for Waze, the popular navigation app. The project focused on identifying users at risk of abandoning the platform and generating targeted recommendations to improve retention.
Cookie Cats A/B Testing
This project analyzes A/B test results from Cookie Cats, a popular mobile puzzle game. The test examined the impact of moving the first gate (where players must wait or pay to continue playing) from level 30 to level 40, focusing on player retention and engagement metrics.
Movie Recommendation System
A hybrid movie recommendation engine with automated data collection. The system scrapes movie data from TMDB and Letterboxd, gathering details like cast, directors, user reviews, and ratings for 300+ movies. Built using Python, the recommendation engine combines content-based filtering with cast and director similarity analysis to suggest movies. Features an interactive Streamlit interface where users can adjust recommendation weights and explore detailed movie information.