Agentic Platform Engineer

Role overview

Help build and scale a platform for AI agents, supporting reliable execution and secure connections to internal data and services. The work spans runtime capabilities, integrations, observability, and deployment tools that let teams develop and operate agents.

Responsibilities

- Build and maintain the agent runtime, including execution, context management, and event streaming. - Create MCP tool servers, APIs, and integrations that provide controlled access to internal services and data. - Develop capabilities such as agent memory, multi-model routing, durable execution, and LLM framework support. - Monitor platform performance, investigate production issues, and improve tracing and evaluation. - Contribute service templates, CI/CD pipelines, and cloud-native deployment patterns for teams building agents. - Add security controls, automated evaluations, and regression tests; own features through deployment and monitoring.

Requirements

- Strong Python skills, including API development, asynchronous programming, testing, and clean architecture. - Practical experience building LLM applications or AI agents, including tool calling, structured outputs, and evaluations. - Understanding of event-driven systems, microservices, Kubernetes, and cloud infrastructure. - Experience with LLM operations, observability, CI/CD, and production troubleshooting. - Knowledge of AI security practices, including authentication, access control, and prompt-injection safeguards. - Ability to own work independently and collaborate with cross-functional teams.

Benefits and work setup

- Remote or hybrid options; remote contracts are available in Spain, Italy, the UK, Belgium, the Netherlands, France, and Germany. - Learning resources and optional language classes. - Home office budget of up to €1,000 and paid trips to the Madrid office.