Job Description

Senior AI Security Engineer | AI Red Teaming / LLM Security

Contract role – Min 6months initially, but very likely extended long term

Start – ASAP, but can wait a notice of up to 30 days

Location – Hybrid Abu Dhabi

We're looking for a highly technical Senior AI Security Engineer to join a major banking organisation in Abu Dhabi, helping to secure enterprise AI platforms, Large Language Models (LLMs) and autonomous AI agents.

This is a genuinely hands-on engineering role, ideal for someone with a strong background in offensive security, application security or penetration testing who has moved into AI/LLM security.

You'll be responsible for identifying vulnerabilities in AI systems, developing and testing security guardrails, and building automated capabilities to protect AI applications and agents.

What you'll be doing

AI Red Teaming: Conduct adversarial testing of LLMs and AI agents, identifying prompt injection vulnerabilities, jailbreaks, unsafe tool execution and data exfiltration risks.
Security Engineering: Design, implement and test AI security guardrails, including input/output filtering, injection detection and agent access controls.
Python Development: Build automated AI security testing frameworks, attack libraries and evaluation harnesses.
Agentic AI Security: Secure AI agents, RAG applications, vector databases, API integrations and MCP-style tool connections.
DevSecOps: Integrate AI security testing into CI/CD pipelines to identify vulnerabilities before production deployment.
Threat Detection: Develop AI-specific monitoring, detection rules and security response capabilities.
Cloud Security: Help secure cloud-based AI infrastructure, with a preference for Microsoft Azure.

What we're looking for

6+ years experience in Security Engineering, Application Security, Offensive Security or Penetration Testing.
At least 1–2 years hands-on experience securing or red-teaming LLMs, Generative AI or agentic applications.
Strong Python programming skills, including developing security automation and testing tools.
Practical experience with prompt injection, jailbreak testing, AI vulnerabilities and adversarial testing.
Understanding of LLM architectures, RAG, AI agents, vector databases and tool/function calling.
Knowledge of OWASP Top 10 for LLM Applications and MITRE ATLAS.
Cloud security experience across Azure, AWS or GCP.
Experience with DevSecOps, CI/CD pipelines and security automation.

Particularly interesting experience