Description

Model Risk Manager | 6+ years | Switzerland | Regulated Financial Services | AI/ML Validation & Python; lead independent second-line validation and governance for AI, machine-learning, generative-AI, and agentic systems, ensuring they remain robust, explainable, fair, secure, and fit for purpose throughout their lifecycle. About the role: We're hiring a Model Risk Manager to extend the model risk management discipline into AI/ML, generative, and agentic systems. You'll act as an independent second line of defense, providing credible challenge to model developers and ensuring models are fit for purpose before and after deployment. This is the right role for someone with a strong quantitative or model-validation background who wants to apply that rigor to the newest generation of AI systems, inside an organization that takes governance seriously. Responsibilities: • Lead independent validation of AI/ML models — including generative and agentic AI — across use cases such as monitoring/detection systems, client analytics, and document and process automation • Provide credible challenge to model developers on methodology, data quality, assumptions, and performance • Assess models for fairness, bias, explainability, robustness, and data privacy risk • Maintain and continuously enhance the model risk framework and model inventory/risk-tiering methodology to cover AI/ML and LLM-based systems • Define and monitor model performance thresholds, ongoing monitoring plans, and re-validation triggers • Prepare validation reports and present findings to model risk committees and senior stakeholders • Track remediation of validation findings and escalate unresolved model risk issues • Support regulatory examinations and audits related to model risk governance, and keep validation standards current with evolving regulatory expectations Experience: • 6+ years in model risk management, quantitative risk, or model validation, with meaningful exposure to AI/ML models • Experience validating or overseeing generative AI/LLM-based systems, or a strong appetite to build this capability • Track record operating an independent second-line function inside a complex, regulated organization • Experience presenting findings to risk committees and senior/executive stakeholders Technical fluency: • Strong quantitative and statistical background, able to critically assess model methodology across classical ML and LLM-based approaches • Understanding of transformer-based/LLM architectures sufficient to evaluate risk — bias, hallucination, explainability — not necessarily to build • Familiarity with model risk frameworks and independent validation tooling and practices • Comfortable with Python and SQL for independent testing and challenge • Exposure to the Azure AI stack and MLOps monitoring tooling, sufficient to assess production model controls Ways of working: • Independent and evidence-based, comfortable holding a firm line under pressure • Strong written and verbal communication, able to translate technical findings for non-technical and executive audiences • Structured, detail-oriented, and rigorous in documentation Qualifications: • University degree in a quantitative field — statistics, mathematics, engineering, computer science, or finance (advanced degree or certification such as FRM, PRM, or CQF a plus) • Fluent English and French; • Eligibility to work in Switzerland Your data: By submitting your resume, you agree to the retention and use of your personal data by TSG for recruitment purposes, including sharing with our clients in the context of your application. The identity and sector of the client will be disclosed to shortlisted candidates ahead of interview.