Senior Platform Engineer – Agentic AI & Harness Engineering

About the role

Kyndryl is hiring a Platform Engineer in Madrid to help build its internal agent harness: the runtime, orchestration and guardrails that turn foundation models into AI agents doing reliable work at scale. You will build and integrate parts of the harness alongside senior platform engineers, with agentic productivity as the goal.

Key responsibilities

• Build and maintain parts of the agent harness: runtime components, tool and MCP integrations, orchestration and guardrail checks.
• Build golden paths and reusable patterns so teams can run agents reliably instead of writing one-off scripts.
• Integrate agents with the systems they act on: CI/CD, infrastructure, cloud services and observability, through APIs and MCP.
• Implement evaluation and guardrail steps so agent behaviour is tested and governed.
• Help track adoption and the productivity impact of agentic workflows.
• Support the teams building and running agents on the harness.

Required profile

• 2-4 years in Platform Engineering, DevOps, SRE or backend software engineering.
• Solid software engineering in Python: production services and automation, not only prompts.
• Professional working English and Spanish.

Required skills

• Cloud-native foundation: Kubernetes, containers, Infrastructure as Code, one major cloud (AWS, Azure or GCP).
• Experience integrating systems through APIs, and an interest in agent and MCP integration.
• An interest in agentic AI and a basic grasp of how LLM agents fail and how engineering makes them reliable.
• Nice to have: agent frameworks such as LangGraph, CrewAI, AutoGen and the Model Context Protocol; AI-assisted development tools (Claude Code, Copilot, Codex); LLMOps, agent evaluation or observability; regulated environments (finance, insurance, public sector).

Career paths are open vertically and horizontally, with opportunities to travel, and Kyndryl offers certifications with Microsoft, Google and Amazon, coaching and hands-on learning.