Remote AI Security Engineer - Agentic Platform

EY; Miami (FL); On-site; USD 140,000 - 180,000; 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 spans from silicon-level attestation and Kubernetes hardening to prompt-injection defense and audit-grade evidence, ensuring defense-in-depth as the platform evolves.

The ideal candidate will have deep cloud-native security experience, be comfortable writing threat models, performing red team activities, and articulating policy-as-code to client CISOs and regulators.

Responsibilities

• Own the platform threat model across deployment targets, including agent autonomy, tool invocation, delegated authority, and multi-tenancy.
• Define the security engineering and control set for every platform layer: infrastructure, Kubernetes, identity, secrets, secure execution, sandboxing, gateway, data, telemetry.
• Set the secure-by-default contract so platform capabilities ship hardened with templates, charts, sandbox profiles, and correct network policies.
• Own defense against agentic threat classes including prompt injection, jailbreak, excessive agency, and retrieval-augmented data exfiltration.
• Work with the architecture team to define the agent authority model, including delegation depth, consent boundaries, and non-escalation invariants.
• Own the sandboxing security standard for agent-generated code execution: isolation, credential scope, egress restrictions, and boundary tests.
• Secure the model and knowledge supply chain: provenance, registry governance, poisoning risk, and integrity of embeddings and vector stores.
• Secure agent-to-agent and tool protocols: discovery trust, tool registration, schema validation, and inter-agent call authorization.
• Lead AI red teaming and adversarial testing to build offensive capability and defense insights.

Job description

EY is seeking an AI Security Engineer to own the security posture of its Agentic AI platform end to end. The role spans from silicon-level attestation and Kubernetes hardening to prompt-injection defense and audit-grade evidence, ensuring defense-in-depth as the platform evolves.

The ideal candidate will have deep cloud-native security experience, be comfortable writing threat models, performing red team activities, and articulating policy-as-code to client CISOs and regulators.