AI Security Engineer: Platform Threat Modeling & Defense

Ernst & Young Oman; Columbus (OH); Remote; USD 150,000 - 210,000; Full time

Job summary

EY is seeking an AI Security Engineer to own the security posture of its Agentic AI platform end to end. The role covers cloud-native security, container orchestration, and secure execution across diverse environments, including highly regulated client settings.

The ideal candidate has depth in platform security, has moved into AI threat models, and can articulate threat models, perform red-team exercises, and defend engineering to CISOs and regulators.

Qualifications

• Genuine depth in cloud-native and platform security.
• Experience moving into AI and agentic threat models.
• Comfort writing threat models, breaking a system with red-teaming, and defending engineering to CISOs.

Responsibilities

• Own the platform threat model across agent autonomy, tool invocation, delegated authority, and data supply chain.
• Define security engineering and controls for infra, Kubernetes, identities, secure execution, network policy, telemetry.
• Set the secure-by-default contract with agent templates, Helm charts, sandbox profiles, and correct controls.
• Lead defense against agentic threat classes including prompt injection, jailbreaks, and authority escalation.
• Define the agent authority model with scoped delegation, consent boundaries, and non-escalation rules.
• Own the sandboxing standard for agent-generated code execution and containment.
• Secure the model and knowledge supply chain including provenance and integrity.
• Secure agent-to-agent protocols and inter-agent calls.
• Lead AI red teaming and adversarial testing to build offensive capability.

Skills

AI security

Threat modeling

Red teaming

Policy-as-code

Job description

EY is seeking an AI Security Engineer to own the security posture of its Agentic AI platform end to end. The role covers cloud-native security, container orchestration, and secure execution across diverse environments, including highly regulated client settings.

The ideal candidate has depth in platform security, has moved into AI threat models, and can articulate threat models, perform red-team exercises, and defend engineering to CISOs and regulators.