Senior AI Security Architect for Agentic Platform

Ernst & Young Oman; Las Vegas (NV); Remote; USD 140,000 - 220,000; Full time

Job summary

EY is seeking an AI Security Engineer to own the security posture of its Agentic AI platform end to end. You will define the platform threat model, implement secure-by-default contracts, and drive defense against complex agentic threat classes.

Ideal candidates have deep cloud-native security experience, leadership in red teaming, and a track record of translating threat models into concrete controls for governance and regulators.

Qualifications

• Experience designing security for AI-enabled platforms and cloud-native environments.
• Ability to articulate and implement threat models for agentic systems and tool invocations.
• Knowledge of Kubernetes hardening, secure boot, and supply chain security.

Responsibilities

• Own the platform threat model across deployment targets and data flows.
• Define security engineering controls for infrastructure, Kubernetes, identity, and telemetry.
• Set secure-by-default contracts for agent templates, Helm charts, and sandbox profiles.
• Lead defense against agentic threat classes including prompt injection and hijacking attacks.
• Define the agent authority model and ensure authority boundaries are enforced.
• Own sandboxing standards for code execution and governance of the knowledge supply chain.
• Secure model provenance, registry governance, and inter-agent communications.
• Lead AI red teaming and adversarial testing to strengthen the platform.

Skills

Platform threat modeling

Cloud-native security

Kubernetes hardening

Secure execution

Threat hunting / red teaming

Tools

Helm charts

Policy-as-code

Threat models

Job description

EY is seeking an AI Security Engineer to own the security posture of its Agentic AI platform end to end. You will define the platform threat model, implement secure-by-default contracts, and drive defense against complex agentic threat classes.

Ideal candidates have deep cloud-native security experience, leadership in red teaming, and a track record of translating threat models into concrete controls for governance and regulators.