This project focuses on building a machine learning pipeline to predict the likelihood of loan default using real-world credit data. The goal is not only to classify applicants as high or low risk, but also to support data-driven decisions around loan approvals using expected loss calculations and a strategy table simulation.