AI Security Engineer

We're looking for an AI Security Specialist to join our client's team and help ensure that AI solutions are developed, deployed, and operated securely and responsibly. In this role, you'll protect AI models, agents, data, and integrations throughout their entire lifecycle while enabling the safe adoption of innovative technologies. You'll combine cybersecurity expertise with a strong understanding of emerging AI risks to design, implement, and test effective security measures, ensuring that human oversight and accountability remain central to every solution. Your Responsibilities
• Conduct threat modelling, risk assessments, and security reviews for generative AI, machine learning, and agentic AI solutions.
• Design and implement security guardrails covering identity and access management, least-privilege principles, secrets management, network isolation, data protection, prompt security, tool usage, output validation, logging, and monitoring.
• Identify and test AI-specific vulnerabilities and misuse scenarios, including prompt injection, data exfiltration, excessive permissions, unreliable outputs, and the misuse of tools or autonomous agents.
• Integrate security requirements into AI sandboxes, gateways, development pipelines, model management processes, dependency management, and production platforms.
• Contribute to security monitoring and incident response strategies for AI components, defining clear ownership, audit trails, monitoring mechanisms, and escalation procedures.
• Evaluate the security and reliability of external AI services, models, libraries, and other dependencies as part of technology and supply chain risk management.
• Translate technical findings and regulatory requirements into practical security measures that support innovation while maintaining an appropriate level of protection.
• Collaborate with IT, security, business, and architecture stakeholders to promote secure AI development and responsible adoption throughout the organisation.

Your Profile
We're looking for a curious and analytical security professional who understands the challenges of emerging AI technologies and can translate complex security risks into effective, pragmatic solutions. Your experience and technical skills include:
• Proven experience in cybersecurity engineering, cloud security, application security, identity and access management (IAM), security operations (SOC), or DevSecOps.
• A strong understanding of AI-specific security risks relating to large language models (LLMs), AI agents, retrieval mechanisms, and model and data supply chains.
• Experience conducting security testing, technical risk assessments, and vulnerability analyses, with the ability to recommend and implement appropriate countermeasures.
• The ability to balance innovation and security by designing proportionate, risk-based controls rather than relying solely on standard checklists.
• Knowledge of relevant security and regulatory frameworks, such as GDPR, the EU AI Act, and NIS2.
• Strong analytical and problem-solving skills, combined with an interest in experimentation and emerging technologies.
• Excellent communication and collaboration skills, with the ability to explain complex technical concepts to both technical and non-technical stakeholders.
• The ability to structure and visualise information, provide clear recommendations, and support multidisciplinary decision-making.
• A results-oriented mindset, with a focus on deadlines, quality, and practical outcomes.
• Experience in public sector environments or specific government cloud and social security contexts is considered an asset. Practical Information
•

Location:
Brussels, with the possibility of working partly from home.
• Languages: Knowledge of Dutch, English, and French is required.
• Application requirement: Candidates are asked to submit a short document of no more than two pages outlining their vision for the safe and responsible use of AI within a public organisation. The document should explain how AI can create tangible value in services and internal operations, with particular emphasis on technical security, the identification of AI-related risks, and the design, implementation, and testing of appropriate safeguards throughout the AI lifecycle. Are you passionate about cybersecurity and eager to help organisations unlock the potential of AI without compromising security? If you enjoy tackling complex challenges and turning emerging risks into practical solutions, we'd love to hear from you Contactperson & Reference
• Reference #: INW28652
• Marilyn Weytens
• marilyn.weytens@i4m.be