AI Security Architect for Agentic Platform Defense

EY; Salem (OR); On-site; USD 180,000 - 240,000; Full time

Job summary

EY is seeking an AI Security Engineer to own the security posture of EY's Agentic AI platform end to end. The role covers threat modeling for agent autonomy, tool invocation, and multi-tenancy across cloud to air-gapped deployments.

Ideal candidates will craft security engineering and control sets for infrastructure, Kubernetes, identity, and secure execution, and define policies, sandboxing, and audit-grade evidence to defend the platform against advanced adversaries.

Responsibilities

• Own the platform threat model: cover agent autonomy, tool invocation, delegated authority, model and data supply chain, multi-tenancy, and deployment targets.
• Define the security engineering and control set for every platform layer: infrastructure, Kubernetes, identity and secrets, secure execution and sandboxing, gateway and egress, data and state, delivery pipeline, telemetry.
• Set the secure-by-default contract so platform capabilities are hardened, including agent templates, Helm charts, sandbox profiles, and network policy.
• Own defense against agentic threat classes: prompt injection, jailbreak, excessive agency, confused-deputy and authority-escalation attacks, tool abuse, data exfiltration.
• Work with architecture to define the agent authority model: delegated authority, scope, consent boundaries, and non-escalation invariants.
• Own the sandboxing security standard for agent-generated code execution: isolation, filesystem scope, egress limits, resource containment, and escape-test suite.
• Secure the model and knowledge supply chain: model provenance, registry governance, poisoning risk, embedding integrity.
• Secure agent-to-agent and tool protocols: discovery trust, tool registration, schema validation, authorization of inter-agent calls.
• Lead AI red teaming and adversarial testing: build the offensive capability and the defense.

Job description

EY is seeking an AI Security Engineer to own the security posture of EY's Agentic AI platform end to end. The role covers threat modeling for agent autonomy, tool invocation, and multi-tenancy across cloud to air-gapped deployments.

Ideal candidates will craft security engineering and control sets for infrastructure, Kubernetes, identity, and secure execution, and define policies, sandboxing, and audit-grade evidence to defend the platform against advanced adversaries.