AI Governance Lead (SDLC & Agentic AI Focus) 
Reports To: Managing Director – AI Governance / Responsible AI
Role Overview As organisations accelerate adoption of Agentic AI and autonomous systems, this role ensures every AI-enabled solution is safe, compliant, and production-ready before deployment.

This is a hands-on AI Governance Lead position embedded directly into the Software Development Lifecycle (SDLC). You will operationalise Responsible AI principles across design, development, testing, deployment, and monitoring — implementing Shift-Left Governance to identify risk before code reaches production.

You will work closely with AI engineering squads, model risk teams, security, and senior stakeholders to integrate Governance-as-Code frameworks, automate compliance controls, and oversee adversarial testing of AI systems.


Core Responsibilities 1. Design & Ideation (Guardrails)

• Lead AI Impact Assessments during the requirements and architecture phase
• Define allowable AI use cases and ethical boundaries
• Ensure models comply with regulatory expectations and Responsible AI principles
• Assess risk in agentic / autonomous AI workflows

2. Development & Coding (Shift-Left Governance)

• Embed automated governance checks into IDEs (e.g., VS Code, Cursor) and CI/CD pipelines
• Implement Governance-as-Code frameworks to reduce manual review
• Scan for hallucinated libraries, insecure AI-generated code, and prohibited data usage
• Integrate automated policy enforcement into MLOps / LLMOps workflows

3. Testing & Validation (Red-Teaming & Adversarial Controls)

• Oversee adversarial testing and red-teaming workstreams
• Conduct jailbreak testing to assess data leakage and prompt exploitation risks
• Validate models for bias, hallucinations, and security vulnerabilities
• Ensure robust Human-in-the-Loop (HITL) controls
• Govern agent-to-agent communication and multi-agent behaviour

4. Deployment & Lifecycle Monitoring

• Establish model drift, bias, and performance monitoring dashboards
• Define kill switches and rollback protocols for production AI
• Ensure audit-ready model lineage, traceability, and observability
• Provide oversight across the full AI lifecycle, from pilot to production

Required Skills & Experience

• Experience in AI governance, model risk, data governance, or technology risk
• Strong technical capability embedding governance into SDLC pipelines
• Hands-on experience with MLOps / LLMOps platforms such as Microsoft Azure Machine Learning or Amazon SageMaker
• Familiarity with adversarial AI testing tools such as Giskard, Deepchecks, or Microsoft Counterfit
• Practical knowledge of the NIST AI Risk Management Framework and ISO/IEC 42001
• Experience governing agentic AI / multi-agent systems
• Ability to translate regulatory and ethical principles into technical enforcement controls

Key Performance Indicators

• Time-to-Trust: Reduce time for AI pilots to pass ethical and risk review
• Governance Automation %: Increase percentage of automated compliance checks within CI/CD
• Risk Mitigation: Zero High or Critical findings in post-deployment AI audits
• Production Resilience: Effective drift detection and safe rollback in live environments