AI Governance Architect

talabat; Dubai; Hybrid; AED 420,000 - 640,000; Full time

Job summary

talabat, the leading on-demand delivery platform in MENA, seeks an AI Governance Lead to define safety, quality, and compliance standards for a growing agent fleet. You’ll craft testing frameworks, risk-tiers, and reusable libraries to scale governance while keeping humans in the loop where needed.

You’ll drive data leakage prevention, tool guardrails, and regulatory mappings, partnering with Legal and Engineering to deliver auditable, compliant AI functionality at pace.

Qualifications

• Extensive experience building and testing safety-critical AI systems.
• Experience with red-teaming or adversarial testing on LLM applications.
• Knowledge of IAM and data access controls; policy-based access is a plus.
• Ability to code in Python to build test harnesses and automation.
• Familiarity with governance, risk, and compliance in regulated environments.

Responsibilities

• Architect adversarial payload libraries and red-teaming frameworks for agent self-tests.
• Author reusable red-teaming templates covering OWASP Top Vulnerabilities for LLMs.
• Enable autonomous data leakage checks and standardized verification suites.
• Provide automated red-team campaigns and regression scripts for safety.
• Develop hallucination testing policies and drift monitoring frameworks.
• Integrate evaluation methodologies into CI/CD pipelines for built-in quality.
• Translate regulatory obligations into actionable technical checklists for teams.
• Maintain governance metrics dashboards and conduct maturity assessments.
• Collaborate with Legal and Engineering to align verification tools.

Skills

Red-teaming

Adversarial testing

Data access control

IAM

Tool guardrails

Python

CI/CD integration

Governance frameworks

Risk assessment

Tools

Python tooling

Job description

talabat, the leading on-demand delivery platform in MENA, seeks an AI Governance Lead to define safety, quality, and compliance standards for a growing agent fleet. You’ll craft testing frameworks, risk-tiers, and reusable libraries to scale governance while keeping humans in the loop where needed.

You’ll drive data leakage prevention, tool guardrails, and regulatory mappings, partnering with Legal and Engineering to deliver auditable, compliant AI functionality at pace.