A dynamic Credit Monitoring System that lowers Loan-Loss Contingency

Create an automated system that helps a commercial bank proactively detect borrower stress before delinquency and take well-informed and timely actions for effective mitigation of credit risk.

• Created EWS framework to strengthen credit risk infrastructure and provide triggers to credit officer for timely actions
• Identified early warning indicators for transactional data, market data, financial data, industry data and other data
• Enabled data sourcing from multiple sources using APIs and Python-based scripts; leveraged machine learning to incorporate news inputs
• Cleaned data to ensure harmonization with internal data for further processing.
• Ran the harmonized data through an EWS engine for validation against predefined rule-based triggers
• Developed support for customizable rules at the user level to generate early warning alerts for timely actions

• Dynamic risk classification of the corporate portfolio
• Early and informed action for credit officers helped lower loan-loss contingency
• Data-driven risk insights to pinpoint stress; information loopback to other functions