The interview discusses the multi-faceted approach to credit risk management within an organization. Credit risk assessment and quantification involve statistical models and non-model methodologies, augmented by qualitative adjustments and expert judgment. The models are created with input from various business units to ensure they meet practical needs and business expectations. This includes using advanced algorithms like machine learning alongside human judgment.

The organization employs a rigorous quality assurance process, conducting independent testing before regulatory review. This helps mitigate risks from faulty models, which could lead to poor decisions and regulatory issues.

For stress testing, the organization adapts models and creates specific scenarios to handle unexpected economic changes, such as those seen during the COVID-19 pandemic.