Through credit risk prediction, this paper investigates how machine learning might enable banks make better lending decisions. We seek to categorize borrowers as either "good" or "bad" credit risks using the IDBI Credit dataset, which comprises information from 1,000 applicants including age, employment status, loan details, and account history. We first carefully explored the dataset and looked for trends that might compromise creditworthiness. We visualized important trends, cleaned and preprocessed the data, and made predictions using several models—including random forests, decision trees, and logistic regression. Our results emphasize which elements most influence a customer's credit risk and show that machine learning can be a useful tool for risk assessment enhancement.