AI security consultant

We are seeking a freelance security expert or specialised security consultancy to perform a comprehensive end-to-end audit in a highly advanced academic R&D setting. You will assess everything from physical access and network architecture to the security implications of AI bias and adversarial robustness of LLMs. You'll also conduct technically advanced DPIAs, translating complex risks into actionable insights.

Information about the organization

The organization is both known for and linked with the most prominent R&D environments of our country. The department features cutting-edge high-performance computing infrastructure, Kubernetes clusters, and AI systems including DGX Basepods, Run:AI and Large Language Models (LLMs).

Deliverables

• Conduct a full security assessment of systems including Kubernetes clusters and AI compute platforms (e.g. DGX B200 / H100)
• Audit NVIDIA Run:AI software stack and related components
• Perform AI bias audits: model architecture reviews, data sampling analysis, and governance evaluation
• Red Teaming of LLMs: simulate attacks, test defences, and advise on mitigation
• Execute complex DPIAs with a focus on ethical, legal, and societal risk analysis
• Produce audit reports with clear, risk-based recommendations
• Share knowledge and best practices with internal teams

Job requirements

• Demonstrated experience with security audits in complex, tech-driven environments
• In-depth knowledge of Kubernetes and AI-focused infrastructure
• Experience with NVIDIA systems (DGX series) and AI pipeline security
• Proven skills in bias audits and red teaming of LLMs
• Strong understanding of DPIA frameworks, including technical and ethical risk analysis
• Excellent analytical and communication skills
• Independent, pragmatic, and proactive mindset
• Based in the EU for timezone and potential travel logistics