[Hiring] AI Security Specialist @UNIQA Group Services

Role Description

This is a position suitable for passionate professionals who want to work for a leading European Insurance Group on a very flexible work schedule. You will work as part of a multinational group with true, large, enterprise-wide systems, complex architectures and real life constraints (speed, performance, non-stop systems availability) and are going to be part of an agile team aimed at designing and implementing enterprise level solutions.

Contract duration: permanent

Schedule: full-time (8 hrs/day, Monday to Friday)

Your responsibilities:

• Define and evolve the security architecture and security controls for the UNIQA Agentic AI Platform across agent runtime, model access, RAG, memory, tool integration, MCP servers, APIs, and cloud infrastructure.
• Design identity and access management for agents, users, services, tools, and tenants using Microsoft Entra ID, managed identities, least privilege, and clear separation of duties.
• Implement controls to detect, prevent, and respond to prompt injection, indirect prompt injection, insecure tool use, data exfiltration, model abuse, excessive agency, and other LLM-specific attack patterns.
• Plan and execute AI threat modelling, security reviews, penetration tests, adversarial testing, and AI red-teaming for platform capabilities and onboarding use cases.
• Define and validate multi-tenant security, country data isolation, network segmentation, secrets management, encryption, and secure connectivity to enterprise systems.
• Establish audit trails, security logging, monitoring requirements, and evidence collection for agent actions, model calls, tool execution, privileged access, and policy decisions.
• Translate security and regulatory requirements into enforceable technical controls and policy-as-code in collaboration with Security, Compliance, Legal, Risk, Data Protection, and Architecture teams.
• Embed security requirements and automated security checks into architecture reviews, CI/CD pipelines, release gates, incident response, and the platform development lifecycle.
• Create reusable security patterns, guardrail standards, reference implementations, and practical guidance for platform, use case, and country teams.
• Assess AI frameworks, models, tools, connectors, and third-party components from a security and supply-chain perspective and document risks and required mitigations.

Qualifications

• Several years of professional experience in cloud security, application security, security architecture, DevSecOps, penetration testing, or comparable cybersecurity roles in complex enterprise environments.
• Proven hands-on experience securing cloud-native platforms, APIs, distributed applications, containers, Kubernetes, identities, networks, secrets, and software supply chains, preferably in Microsoft Azure environments.
• Advanced knowledge of IAM and Microsoft Entra ID, including OAuth2/OIDC, managed identities, workload identities, role-based access control, privileged access, and tenant isolation.
• Practical understanding of generative AI and agentic AI security risks, including prompt injection, tool and plugin abuse, sensitive data disclosure, model extraction, data poisoning, insecure output handling, excessive agency, and denial-of-service scenarios.
• Experience with structured threat modelling, security design reviews, penetration testing, adversarial testing, AI red-teaming, vulnerability management, and risk-based remediation.
• Familiarity with relevant AI security guidance and threat catalogues such as the OWASP Top 10 for LLM Applications, MITRE ATLAS, NIST AI Risk Management Framework, or comparable industry practices.
• Experience implementing policy-as-code, automated security testing, security gates, container and dependency scanning, and traceable controls within CI/CD and GitOps processes.
• Good understanding of logging, monitoring, auditability, incident response, and forensic requirements for AI-driven and highly automated systems.
• Ability to translate GDPR, EU AI Act, DORA, and internal security requirements into technically precise, testable controls together with the accountable specialist functions.
• Strong analytical thinking, security ownership, pragmatic risk communication, documentation discipline, and the ability to collaborate with engineering, architecture, operations, compliance, and international country teams in English.

Benefits

• Very stable work environment, as part of a large Insurance Multinational Group.
• Flexible work program, respecting your own private time.
• Attractive remuneration package.
• Lunch tickets.
• Special Days vouchers (Women’s Day, Children's Day, Christmas, Easter).
• Private pension and private health insurance contribution.
• Time off when holidays fall on a weekend (up to 3 days/year).
• Exchange of experience and training with international professionals, as a premise for personal development.
• Team building events.