Senior AI Security Architect & Platform Defense Lead

EY; San Mateo (CA); Remote; USD 140,000 - 210,000; 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, secure-by-default controls, sandboxing, and defense against agentic threat classes across cloud to air-gapped deployments.

The candidate will lead defense against prompt injection, jailbreaks, and authority escalation, while defining the agent authority model and ensuring end-to-end supply-chain integrity and policy enforcement.

Responsibilities

• Own the platform threat model: cover agent autonomy, tool invocation, delegated authority, model and data supply chain, multi-tenancy, and deployments from cloud to air-gapped.
• Define the security engineering and control set for every platform layer: infrastructure/boot chain, Kubernetes/cluster fabric, identity/secrets, secure execution and sandboxing, gateway/egress, data/state, delivery pipeline, telemetry.
• Set the secure-by-default contract so platform capabilities arrive hardened, including agent templates, Helm charts, sandbox profiles, and network policy.
• Own defense against agentic threat classes: prompt injection, jailbreaks, excessive agency, authority escalation, tool/function-call abuse, memory/context poisoning, data exfiltration.
• Work with architecture to define the agent authority model: delegated/null authority, scope/delegation-depth limits, consent boundaries, non-escalation invariants.
• Own the sandboxing security standard for agent-generated code: isolation, filesystem/credential scope, egress restrictions, resource containment, escape-test suite.
• Secure the model and knowledge supply chain: provenance, registry governance, poisoning risk, embedding/vector-store integrity.
• Secure agent-to-agent and tool protocols: discovery trust, tool registration, schema validation, inter-agent authorization.
• Lead AI red teaming and adversarial testing: build offensive capability and cadence to test guardrails, sandboxes, and boundaries.
• Own supply-chain integrity end-to-end: artifact signing/verification, SBOM generation/attestation, provenance and license governance.
• Own admission and runtime policy: policy-as-code across Kyverno/OPA, signature verification, Pod Security Standards, enforce guardrails.

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, secure-by-default controls, sandboxing, and defense against agentic threat classes across cloud to air-gapped deployments.

The candidate will lead defense against prompt injection, jailbreaks, and authority escalation, while defining the agent authority model and ensuring end-to-end supply-chain integrity and policy enforcement.