My colleague and I built this project because we were curious about how city complaints and neighborhood conditions actually affect real estate value in NYC. Instead of just analyzing a static dataset, we wanted to create something interactive — a system where we could store, process, and explore both 311 complaints and property sales in one place. So we designed a full-stack platform using SQL, Python, and Flask that lets us analyze trends, compare neighborhoods, and visualize how complaint volume correlates with housing prices.

As someone with interest in finance and personal banking, I wanted to explore how machine learning could help predict loan approvals based on customer data. I gathered a dataset of bank customers, cleaned and preprocessed the data using Pandas, and then built classification models with scikit-learn to predict whether a loan would be approved or denied. I evaluated different algorithms, tuned hyperparameters, and visualized the results to understand which factors most influenced loan decisions. This project helped me apply data science techniques to a real-world financial problem while deepening my understanding of ML workflows.