- Business Understanding: Collaborated with stakeholders to understand business operations and align the AI model with the bank's risk management strategies. Conducted exploratory data analysis (EDA) to validate insights and ensure they were consistent with business objectives. - Data Integration and Preprocessing: Conducted data profiling, distribution checks, and handled missing data and outliers to ensure high-quality input for the model. Integrated data from multiple sources, including credit bureaus and financial statements. - Feature Engineering: Selected relevant features through methods like R2 value from DecisionTreeRegressor, optimizing the model's predictive power. - Model Development: Designed and implemented a machine learning model to predict the probability of default and other risk factors using Logistic Regression, Random Forest, and XGBoost. - Model Validation: Performed model validation using techniques such as K-Fold Cross Validation and back-testing to ensure robustness and accuracy. - Deployment and Integration: Successfully deployed the model into the bank’s existing loan origination system, enabling real-time decision-making in the credit underwriting process. Impact: ------------------------------------------------ - The model reduced loan default rates by accurately identifying high-risk applicants, leading to a 6% increase in overall profits.Improved processing efficiency, reducing underwriting time, and enhancing customer satisfaction. - Enhanced the bank's ability to manage credit risk, resulting in more sustainable lending practices