IT Security Penetration Tester - Threat Modeling activities (42785)
Summary
Senior IT Security Penetration Tester focused on threat modeling for AI systems, including LLMs and machine learning agents. The role involves leading workshops, identifying attack surfaces using STRIDE and kill chains, and mapping AI-specific threats to security controls and business risks.
We require a senior security specialist with practical experience in AI-related threats and a strong understanding AI related threats. The specialist will independently conduct application-level assessments, map relevant AI threat scenarios, evaluate security exposures, and define appropriate risk mitigation strategies.
Responsibilities
Lead threat modelling workshops with application and platform teams.
Identify and assess assets, trust boundaries, attack surfaces, threats, and security risks related to AI threats.
Develop and maintain threat models using the approved threat modelling tooling and methodology.
Review solution designs and architecture documentation to identify security gaps and validate threat coverage.
Map applicable AI threat scenarios to application components, data flows, and security controls.
Collaborate with development, architecture, and security teams to define practical risk mitigation actions.
Provide guidance to less experienced team members during threat modelling activities.
Track and report identified threats, security risks, mitigation actions, and remediation status.
Practical experience in threat modelling AI agents, machine learning systems, and large language models.
Strong working knowledge of threat modelling methodologies, including STRIDE, attack trees, and kill chains.
Ability to independently identify assets, trust boundaries, attack surfaces, threat actors, and abuse cases.
Experience developing and adapting reusable threat models and threat libraries.
Ability to map identified threats to relevant security controls, risks, and remediation actions.
Strong working knowledge of the OWASP Top 10 for LLM Applications and MITRE ATLAS, with the ability to apply these frameworks during threat assessments.