Founding Engineer — Agent Control Plane

Guild.ai • San Francisco, CA

7h

About The Position

We're building the control plane to support the use of AI agents in teams and companies.
As a Founding Engineer focused on the Agent Control Plane, you will design and build the systems that run, manage, secure, and govern AI agents operating in real-world production environments.
You’ll work on distributed systems, auth, security boundaries, infrastructure orchestration, and deep integrations with modern foundation models. You’ll help define how organizations safely deploy and control complex, multi-agent systems at scale.
If you’re excited about building the infrastructure layer that makes agentic software safe, reliable, and enterprise-ready - we'd love to talk!

Requirements

• Experience building distributed backend systems
• Deep understanding of authentication & authorization systems
• Experience with Kubernetes and cloud-native infrastructure
• Experience integrating with LLM APIs or building agentic systems
• Strong Python and/or TypeScript proficiency
• Comfort designing multi-tenant SaaS systems

Nice To Haves

• Experience building developer platforms
• Experience with security-critical systems
• Experience in compliance-aware environments (SOC2, ISO, etc.)
• Experience with GitHub/Slack/Enterprise API integrations
• Background in infrastructure tooling or control planes

Responsibilities

• Build core infrastructure that enables:
• Agent Runtime & Orchestration
• Systems for running and managing autonomous agents
• Multi-agent coordination and lifecycle management
• Observability, tracing, evaluation, and reliability tooling
• State management and memory systems
• Tooling & Integrations
• Secure execution of agent tools (e.g., GitHub, Slack, internal APIs)
• Tool abstraction layers and permission boundaries
• Sandboxed and policy-controlled tool invocation
• Extensible plugin systems
• Authentication & Authorization
• Fine-grained, role-based access control (RBAC) and policy systems
• Enterprise-grade auth (OAuth, SAML, SCIM, API keys)
• Multi-tenant architecture
• Secure identity propagation between users, agents, and tools
• Security & Compliance
• Agent permission scoping and guardrails
• Data isolation and tenant separation
• Audit logging and traceability
• SOC2-minded system design
• Secure execution environments
• Cloud Infrastructure
• Deploying and operating systems on GCP, AWS, and Azure
• Kubernetes-native systems (GKE today)
• Infrastructure as Code
• Scalable, resilient backend services
• Foundation Model Integration
• Deep integration with foundation models from OpenAI, Anthropic, and Google
• Model routing and abstraction layers
• Cost control and performance optimization
• Structured tool calling and agent frameworks

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed