Security Engineer: AI-Agent Platforms & Zero-Trust Policy

Intellias; United States; Remote; USD 150,000 - 190,000; Full time

Job summary

Intellias is seeking a Security / Policy Engineer to design and implement authorization models for AI agents and cloud-native platforms. The role focuses on Cedar, OAuth/JWT-based authorization, and zero-trust security.

You will work with enterprise security teams to define principals, resources, and conditional access, develop policy-as-code, and ensure auditability of policy decisions across distributed systems.

Qualifications

• Hands-on experience designing ABAC models and least-privilege access.
• Strong background in cloud security and identity management.
• Familiarity with IAM governance and enterprise security reviews.

Responsibilities

• Design and implement authorization policies using Cedar and AWS AgentCore Policy.
• Develop policy-as-code frameworks for cloud-native and AI-agent platforms.
• Manage ABAC models with least-privilege and default-deny principles.
• Integrate authorization with OAuth 2.0, JWT, and Microsoft Entra ID.
• Define principals, resources, actions, and conditional access policies.
• Define token validation, claims mapping, and authorization workflows.
• Implement audit logging and traceability for policy decisions.
• Collaborate with security and architecture teams on governance.
• Contribute to zero-trust strategy across distributed systems.
• Evaluate AWS authorization tech including Cedar and AWS Verified Permissions.

Skills

Cloud security

Identity engineering

ABAC design

Zero-trust architecture

Policy-as-code

Tools

Cedar (AWS AgentCore Policy)

OAuth 2.0 / JWT

Microsoft Entra ID

Policy-as-code tooling

Job description

Intellias is seeking a Security / Policy Engineer to design and implement authorization models for AI agents and cloud-native platforms. The role focuses on Cedar, OAuth/JWT-based authorization, and zero-trust security.

You will work with enterprise security teams to define principals, resources, and conditional access, develop policy-as-code, and ensure auditability of policy decisions across distributed systems.