5 months ago
Remote, United StatesSenior
Responsibilities
- Architect and implement authentication and authorization systems for MCP servers, including OAuth 2.0, Dynamic Client Registration, and token management.
- Build and extend the OAuth broker for enterprise identity integrations across multiple vendors.
- Design secure, auditable identity propagation for AI agents.
- Integrate with enterprise identity providers and SCIM systems.
- Define fine-grained access control policies for MCP tools and resources.
- Collaborate directly with enterprise customers to solve identity and security challenges.
- Contribute to the MCP Auth specification and help define agent identity standards.
Requirements
- 5+ years of software engineering experience with significant focus on identity, authentication, or authorization systems.
- Deep experience with OAuth 2.0 and OIDC, including Dynamic Client Registration, token exchange, and audience restriction.
- Experience building or integrating with enterprise identity systems such as Okta, WorkOS, Auth0, AWS IAM, or GCP IAM.
- Strong distributed systems and API security fundamentals.
- Experience with Python and TypeScript, using a Python/FastAPI backend and TypeScript/React frontend.
- Comfort working directly with enterprise customers to understand and solve security requirements.
- Experience with Kubernetes-native authorization patterns or service mesh security is preferred.
- Background in ML security, including differential privacy or LLM security research, is preferred.
- Experience with identity for multi-tenant SaaS platforms is preferred.
- Familiarity with the MCP specification is preferred.
- Proficiency with AI-assisted development tools such as Claude Code or Cursor is expected.
Benefits
- Competitive salary and equity package.
- Four weeks of paid vacation, paid sick leave, and paid parental leave.
- Professional development budget for conferences, courses, and certifications.
- Choice of laptop and accessories.
- Comprehensive health, dental, and vision coverage.
- Opportunities to work directly with enterprise customers.
Tech Stack
About Runlayer
Runlayer builds an enterprise MCP platform that securely connects AI agents, skills, and tools to internal systems with governance, permissions, and observability. It sells its platform to enterprises adopting AI across functions, focusing on security and control in complex environments. The privately held company is headquartered in New York, with customers including Gusto, Instacart, Opendoor, dbt Labs, and Decagon.