AI Security Engineer - Agentic AI, AI Hubs & Use Cases

Job Overview

As an AI Security Engineer, you are the first-line security engineering role within the Central AI team / AI Hubs. Your mission is to ensure that AI use cases, agents, and supporting platform components are built and operated securely by design and in line with the groupwide security framework.

You will work hands-on with AI product, engineering, and platform teams to translate central security requirements into practical technical controls, implementation patterns, evidence, and delivery guardrails. The role involves close collaboration with the Group CISO department to ensure alignment with group standards, security architecture principles, and control expectations.

This is a hands-on role with strategic weight. You will work closely with Group AI leadership, the Agentic Platform team, Group CISO, the Data Protection Office, and country IT and security leads.

Main Responsibilities

• Embed security by design across AI use cases, agents, and platform integrations, with controls for access, secrets, logging, monitoring, sandboxing, tool and API restrictions, and external integrations.
• Translate group security policies and architecture into practical controls, reusable patterns, and guidance for AI teams, ensuring alignment with risk, third-party, supplier, and operational requirements.
• Prepare security-gate evidence, drive remediation from reviews, and escalate unresolved control gaps, deviations, or delivery risks.
• Address AI-specific risks, including prompt injection, data leakage, and insecure tool use through hardening and secure deployment patterns.
• Support due diligence and secure implementation for external AI vendors, models, orchestration components, and APIs.
• Strengthen AI incident readiness through detection, logging, response scenarios, playbooks, exercises, and collaboration with SOC and central security teams.
• Scale secure AI delivery through reusable patterns, stronger hub capability, and lessons shared back into the groupwide security model.

Key Requirements

• 7+ years in IT and cybersecurity, with experience in transformation, digital, or platform contexts.
• Solid technical grounding in cloud security, identity and access management, application security, and secure SDLC.
• Working knowledge of AI security, including prompt injection and data leakage risks.
• Experience with threat modeling, risk assessment, and security architecture for distributed systems.
• Strong stakeholder management and ability to influence solution designs constructively.
• Pragmatic, outcome-driven, and resilient under ambiguity.
• Fluent in English; an additional European language is a plus.

Nice to Have

• Prior security responsibility in a multi-country, federated environment, such as logistics, retail, or financial services.
• Hands-on exposure to LLM and agent platforms.
• CISSP, CISM, or equivalent.
• Experience contributing to or running AI red-teaming exercises.

Other Details

This role sits at the center of the AI agenda, ensuring that AI agents are safe, data is protected, and controls enable countries to deliver fast without inheriting unmanaged risk.