Director - NFR Model Validation

Job Summary

• Support GMV Head in performing an independent validation of new and existing models that are used in Non-Financial Risk areas such as Financial Crime Risk, Fraud Risk, Compliance Risk, Operational Risk, etc.
• This includes but not limited to review of model development process including underlying assumptions & theoretical basis, quantitative assessment of model performance, documentation of validation finding and communicate to senior management in relevant forum.

Key Responsibilities:

• Support GMV senior manager in performing an independent validation of new and existing models that are used in Non-Financial Risk management pertaining to Financial Crime Risk, Fraud Risk, Compliance Risk, Operational Risk, etc of bank.
• Define and drive strategic NFR goals in CoE-India aligned with larger SCB's risk strategy and risk appetite.
• Work with senior leadership on managing local attrition and hiring (if required).
• Complete ownership of delivery around model submitted for validation providing independent challenge, adherence to regulatory expectation, accountable for quality, timeliness and effectiveness of deliverables.
• Engagement with senior stakeholders from different lines of defence mainly, first line of defence and third line of defence (Audit) to manage issues, governance and closure.
• Qualitative review of model development process including underlying assumptions & theoretical basis.
• Quantitative assessment of model performance via data evaluation and statistical testing.
• Thorough evaluation of AI Risks for those models which use AI in the processing component.
• Documentation of validation findings and communication of results to senior management and presentation to relevant committees.
• Review regulatory requirements and industry practice regarding the models.
• Assist Head of Model Validation in addressing concerns or questions relating to the models.
• Lead implementation of newer tools and technology around NFR validation to bring in process efficiency.
• Accountable for delivery of all validation completed in CoE-India providing high-quality review of validation completed by members of NFR team, guide the team in resolving issues related to stakeholders, strategic communication both with senior line of management and junior members of the team.
• Anticipate and act proactively on any challenges around efficient delivery of model validation.

Skills and Experience:

Must-have Skills:

• Drives Vision & Purpose
• Gives Clarity and Guidance
• Strategic Mindset

Other Skills:

• Collaborates
• Courage
• Nimble Learning
• Decision Quality

Motivations:

• Financial crime models play a critical role in identifying suspicious activity, detecting potential money laundering and helping the bank meets its regulatory compliance. Any breach in such models leads to not only huge financial loss but also reputational loss.
• This role requires "thinking beyond model", not just its statistical performance.
• This role requires ability to "Challenge with confidence". The associate should have an investigative mindset.

Qualifications:

• Key Experiences: Previous experience in model validation, model development, quantitative analysis, or risk management roles, preferably at least 10 years, with a reasonable focus on AI risk.
• Certifications/Qualifications: A bachelor's or master's degree in a quantitative field such as Artificial Intelligence, Machine Learning, Data Science, Statistics, Finance, or a related discipline.
• Language: SAS, Python, R
• Proven experience in developing and implementing risk management frameworks and methodologies.
• Strong understanding of Financial Crime Risk and Fraud Risk landscape.
• Strong knowledge of relevant regulations (i.e., EU AI Act, SS 1/23, SR 11-7), industry best practices related to AI risk, and model risk management frameworks.
• Good understanding of model life cycle management and model risk management.
• Thorough understanding of Data Science, Machine Learning and GenAI models including knowledge of frameworks and techniques.
• AI/ML Fundamentals: Familiarity with algorithms, techniques, calibration, evaluation and testing.
• Supervised, unsupervised learning models (such as classification, regression, clustering) and reinforcement learning.
• Deep learning concepts including GenAI such as LLMs and the respective terminologies, i.e., LLM architecture, prompt engineering, finetuning, Retrieval Augmented Generation (RAG), evaluation framework and benchmark.
• Evaluation of model performance including accuracy, stability, fairness.
• Ability to address challenges of AI models; black-box nature, drift (data drift, concept drift, model drift), and ethical considerations.
• Explainability and Interpretability: Demonstrated knowledge and practical experience with techniques for model explainability and interpretability, such as LIME and SHAP to address black-box nature of models.

Skills

Model Validation, Python, R, Risk Management, SAS