AI Security Engineer

Description

The company's Digital Identity & Security unit provides software, biometric and encryption technologies and services that help organizations and governments verify identities and keep data safe.

This role sits between strategy and delivery. The AI Security Engineer builds, ships and maintains AI systems for client projects and internal needs, with security built in from the design stage.

The AI Security Advisor sets strategy, governance and advisory direction, and you turn that direction into working solutions.

You combine software engineering with interest in applied AI (LLMs, agents, RAG, MLOps) and the security risks that come with it. Most of your time goes to client work. The rest goes to internal initiatives such as AI-enabled business tools, agentic security automations and shared RAG platforms, which also serve as a learning ground and let you reuse what you learn from clients.

Responsibilities

• Build client solutions that need software engineering and secure AI integration, with autonomy matched to your profile and the engagement.
• Join scoping workshops to understand the business need, turn it into a technical architecture and identify security and compliance constraints.
• Design, prototype and develop LLM integrations, RAG, agents and AI-driven automations within the framework set by the AI Security Advisor.
• Take client projects through to production, covering industrialization, secure deployment pipelines and handover to the client's internal teams.
• Support pre-sales with demos, PoCs, technical estimates and technical RFP sections, working alongside sales and the AI Security Advisor.
• Present your work to clients clearly, teaching and advising as you go.
• After each engagement, record lessons learned and reusable components, and feed them into internal frameworks and standards.
• Contribute to internal AI-enabled business tools, such as CRM, due diligence or decision support, applying quality and security practices.
• Maintain document processing pipelines, keeping them robust and protecting the data they handle.
• Build agentic components for cybersecurity operations, especially SOC use cases, working with the Cyber Defense teams.
• Create RAG, agent and automation tools for IT, HR and other departments.
• Turn PoCs and PoVs from the AI team into robust, maintainable solutions, then document, test and maintain them over time.
• Write code that follows secure coding principles, for client work and internal work alike, and apply application security checks throughout.
• Harden the pipelines that deploy AI, covering CI/CD, secrets handling, dependency checks, and API and cloud links.
• Add defenses for AI systems such as guardrails, input and output filtering, monitoring, and control over the tools agents can use.
• Work with the AI Security Advisor to apply OWASP LLM Top 10, MITRE ATLAS and NIST AI RMF in projects, and take part in threat modeling and code reviews of AI components.
• Track new tools, frameworks and practices in AI development and application security.
• Build reusable libraries, scripts and components that speed up future projects.
• Share knowledge through lessons learned, demos and documentation, and feed project insights into the standards the AI Security Advisor leads.
• Keep execution aligned with strategy by working closely with the AI Security Advisor, delivery teams and R&D.
• Explain technical topics clearly to developers, DevSecOps and MLOps staff, and business topics to business and management audiences.
• Help coordinate with the company's global AI initiatives.

Requirements

• Strong software engineering basics, ideally in Python, with clean, tested and maintainable code.
• Curiosity about modern AI architectures and some hands-on experience with LLMs, RAG, agent frameworks or API integrations.
• Understanding of common web vulnerabilities (OWASP Top 10) and secure coding principles.
• Foundations in one cloud platform (AWS, Azure or GCP) and in DevOps, including CI/CD, containers and infrastructure as code.
• Skills you will grow into: production-grade AI, including advanced RAG, agents, orchestration and observability.
• Knowledge of AI-specific risks such as prompt injection and data poisoning, and of defenses such as guardrails, agent sandboxing and tool access control.
• Familiarity with OWASP LLM Top 10, MITRE ATLAS, NIST AI RMF and the AI Act.
• Practice in MLOps and in securing AI deployment pipelines.
• Real passion for AI and cybersecurity, shown through side projects, open source work, experiments, blogging or regular reading.
• Good interpersonal skills and a willingness to work directly with clients: listen, rephrase, simplify and defend a technical position.
• Curiosity, autonomy and fast learning in a field that changes every week.
• A delivery mindset: finish what you start, ship working things and iterate.
• Clear communication in French and English, written and spoken, for technical and business audiences.
• Eagerness to work closely with the AI Security Advisor and to grow in AI, security and consulting.
• The role is open to all experience levels, from junior to senior, and scope, autonomy and client-facing work will be set to match.
• The common base is a software engineering background plus a real interest in cybersecurity, AI and consulting.
• For junior candidates, passion, fast learning and appetite for project work count more than formal experience. For senior candidates, prior client work and AI or application security projects are valued.
• Personal projects in AI, software engineering or application security are a strong plus at any level.
• Native French or Dutch speaker.
• English is a plus.
• No certification is needed at hiring. A skills plan may include GIAC AI security certifications.
• GIAC AI Platform Security (GAIPS): securing GenAI applications and LLM pipelines, including integrations, APIs, MLOps and agents.
• GIAC AI Security Automation Engineer (GASAE): automation and AI for offensive, defensive and cloud security, especially agentic cybersecurity automation.
• GIAC Offensive AI Analyst (GOAA): offensive AI techniques such as deepfakes, guardrail bypass and AI-assisted malicious code, useful for informing defense.
• Other cybersecurity or cloud certifications, such as OSCP, OSCE or AWS, Azure or GCP certifications, are also valued.

Conditions

Reports to the AI Security Advisor and senior management.

The company is a disability-committed employer and values diversity.