AVP, Credit Risk Data Scientist, Credit Risk Modelling

What matters most

• Model Development and Lifecycle Management

  The core responsibility is to develop, implement, and maintain machine learning models for credit risk across diverse portfolios including Consumer, Small Business, and Wholesale. This requires not only building predictive models but also ensuring their ongoing performance through rigorous back-testing, monitoring, and reporting to detect weaknesses early.

• Regulatory Compliance and Model Governance

  The role requires strict adherence to governance frameworks. You must collaborate effectively with independent model validators to address findings and engage directly with auditors and regulators to ensure all credit risk models meet institutional and external compliance standards.

• Stakeholder Engagement and Business Integration

  You will act as a bridge between technical model outputs and business strategy. This involves translating complex analytical solutions into actionable insights for credit decisioning, allowance, and capital assessment, requiring the ability to communicate effectively with senior management.

Likely challenges to think through

• Balancing Innovation with Strict Governance

  Integrating advanced machine learning techniques into a highly regulated banking environment often creates tension between model complexity and the need for interpretability and compliance. Prepare by demonstrating how you have successfully navigated model validation processes while maintaining the integrity of sophisticated predictive models.

• Managing Diverse Portfolio Requirements

  The role spans Consumer, Small Business, and Wholesale segments, each with distinct risk profiles and data characteristics. You may face difficulty in standardizing model approaches across these segments; prepare by highlighting your experience in adapting methodologies to suit different credit risk contexts.

• Technical Complexity in Big Data Environments

  Working with large-scale datasets using Hadoop, Hive, Trino, and Spark requires significant technical overhead. The challenge lies in maintaining model performance and scalability within these systems; be prepared to discuss how you optimize data pipelines and model deployment in a DevOps-driven environment.

Topics to prepare

• Credit Risk Modelling Frameworks

  Review IFRS 9 Expected Credit Loss (ECL) methodologies, internal rating models, and credit stress testing. This is essential as the role involves managing these specific model types within the Group Risk Management team.

• Model Governance and Validation

  Study regulatory requirements and internal model governance frameworks. The role requires collaborating with independent validators and engaging with auditors/regulators to ensure compliance and address validation findings.

• Model Performance Monitoring and Back-testing

  Understand statistical techniques for back-testing and monitoring model performance. This is a core responsibility to ensure adherence to performance standards and early detection of model weaknesses.

Skills to revise

• Machine Learning Frameworks (Scikit-Learn, Tensorflow, PyTorch)

  Practice implementing predictive models using these libraries. The role explicitly requires proficiency in these tools to develop and maintain machine learning credit risk models.

• Big Data Technologies (Hadoop, Hive, Trino, Spark)

  Refresh knowledge on querying and processing large datasets using these technologies. You will be expected to collate and analyze large-scale data for credit risk assessment.

• DevOps and Version Control (Git, Jenkins, Jira)

  Familiarize yourself with CI/CD pipelines and project tracking tools. These are required for managing model configurations, parameters, and the development lifecycle within the team.

• Prepare a technical project walkthrough

  Select a past project where you developed a predictive model for credit risk or a similar financial application. Be prepared to explain your choice of algorithms (Scikit-Learn, Tensorflow, or PyTorch), how you handled large datasets using big data tools like Spark or Hive, and how you validated the model's performance.

• Review credit risk modelling methodologies

  Refresh your knowledge on industry-standard frameworks, specifically IFRS 9 Expected Credit Loss (ECL) models, internal rating models, and stress testing. Be ready to discuss how machine learning can be integrated into these traditional risk management structures.

• Demonstrate model governance and compliance experience

  Prepare examples of how you have collaborated with independent model validators or auditors. Focus on your process for documenting model parameters, addressing validation findings, and ensuring your work complies with regulatory standards.

• Practice explaining complex models to non-technical stakeholders

  Develop a concise way to communicate the business impact of your models—such as improvements in credit decisioning or capital assessment—to senior management. Focus on translating technical performance metrics into actionable business insights.

• Formulate strategic questions for the interviewers

  Prepare thoughtful questions regarding the team's current data infrastructure, the specific challenges they face in model back-testing, or how the team balances innovation with the bank's strict model governance frameworks.