About the Role
Build backend systems and infrastructure that make agent execution reliable. Working with the technical leadership and AI engineering team, you will own foundational architecture for persistent state, safe execution, permissions, and recovery across humans, agents, and external systems.
What You'll Do
Build backend services and data models to track operational work, ownership, and outcomes over time.
Develop asynchronous workflows that reliably handle retries, timeouts, external events, and interrupted execution.
Create shared APIs and tool-execution interfaces with authorization and customer-data isolation.
Improve how operational knowledge is stored, versioned, and served.
Strengthen testing and observability as the platform evolves.
Investigate production problems, design clear interfaces, and deliver maintainable solutions.
What We're Looking For
At least 3 years of backend engineering experience building and maintaining production services, with experience relevant to production agent infrastructure.
Practical Python and/or TypeScript/Node.js experience, and willingness to work across both.
Experience with PostgreSQL or another relational database, including data modeling and transactions.
Experience with background jobs, queues, or event-driven processing, including failure handling and data consistency.
Ability to build resilient asynchronous workflows and investigate production issues.
Experience with workflow engines, multi-tenant applications, authorization, FastAPI, or agent tooling is a plus. Prior LLM experience is not required.
Compensation & Benefits
€80,000 annually. Visa sponsorship is not available.
Location
On-site in Berlin, Germany.