Role Overview

The Data & AI Governance Lead will design, implement, and operationalize enterprise-grade data and model governance frameworks to support AI, ML, and GenAI initiatives within regulated environments. The role ensures that AI solutions are secure, compliant, auditable, and scalable, while enabling innovation across business stakeholders. The position partners closely with Data Management Office (DMO), Risk, Compliance, Finance, IT, and business teams to translate regulatory expectations into practical and enforceable controls across Azure, Databricks, and AI platforms.

Key Responsibilities

1. Data Governance for AI

• Define and operationalize enterprise data classification, metadata standards, and attribute tagging for AI usage, including PII/PCI/PHI, sensitivity tiers, usage restrictions, and consent.
• Design access and entitlement models for AI and agent-based workloads, including:
  
  • Azure Entra ID integration
  • RBAC/ABAC entitlement models
  • Privileged access management
  • Break-glass procedures
  • End-to-end audit trails and traceability
• Establish metadata, catalog, and lineage operating models using Collibra and/or Microsoft Purview, covering:
  
  • Lineage for AI pipelines
  • Governance of API-based data acquisition
  • Governance of vector databases and RAG stores
• Define cross-border data usage rules aligned with MAS, HKMA, JFSA, GDPR, PDPA and other regulatory frameworks.
• Set governance boundaries for platforms such as Databricks and Azure, including:
  
  • Unity Catalog governance
  • Lakehouse permissions and Delta Sharing
  • Data residency and jurisdictional controls

2. Model Governance (LLM / ML)

• Implement lifecycle governance across ML and LLM models, including:
  
  • Model inventory and registration
  • Risk classification and tiering
  • Approval workflows and sign-offs
  • Model documentation and validation standards
  • Human-in-the-loop controls and checkpoints
• Establish observability and monitoring for ML/LLM models, including:
  
  • Drift, bias, and performance monitoring
  • Prompt and output logging
  • Toxicity detection and filtering
  • Rollback mechanisms and release governance
  • Integration with MLflow and Model Registry platforms
• Operationalize AI safety, including:
  
  • Red-teaming and adversarial testing
  • Secure prompt design principles
  • Sensitive data minimization and retention
  • AI incident response and escalation procedures

3. Ways of Working & Operating Model

• Define and maintain AI governance frameworks, standards, policies, procedures, and RACI models.
• Partner with DMO, Risk, Compliance, Finance, and IT stakeholders to ensure regulatory alignment and risk coverage.
• Translate governance principles into technical controls across Azure cloud services, Databricks Lakehouse, LLM platforms, and AI agents.
• Serve as a trusted advisor, balancing regulatory expectations and innovation enablement.

Candidate Profile (Must Have)

• 10+ years of experience leading data governance, AI governance, or model governance in regulated industries (preferably financial services).
• Hands-on experience with:
  
  • Azure (including Entra ID)
  • Databricks and Unity Catalog
  • MLflow/Model Registry
  • Collibra and/or Microsoft Purview
  • Enterprise IAM and access control models
• Strong understanding of APAC regulatory expectations and cross-border data requirements.
• Proven experience developing policies, standards, and operating procedures.