Senior Platform Engineer – Agentic AI & Harness Engineering
Hybrid · Madrid, Madrid, Spain
Job Summary
Senior Platform Engineer to design and build the agent harness (runtime, orchestration, tool and MCP integration, context and memory, evaluation, guardrails and observability) and define the oversight model (policy-as-code, human-in-the-loop checkpoints, loop and failure detection, evaluation harnesses, cost control). Role involves integrating agents with CI/CD, infrastructure, cloud services, observability and ticketing via APIs and MCP; create measurable agentic productivity, govern harness upgrades across model/provider changes, and mentor teams to reduce friction between Dev, Ops, Security and users. Requires hands-on Python development, strong cloud-native experience (Kubernetes, IaC, AWS/Azure/GCP), understanding of LLM agent failure modes, platform-as-a-product mindset (contracts, SLOs, adoption metrics), and bilingual English/Spanish communication. Nice-to-have: agent frameworks (LangGraph, CrewAI, AutoGen), Model Context Protocol, and experience with AI-assisted development tools and observability for agentic systems.
Required Qualifications
- 5+ years in Platform Engineering, DevOps or SRE, building developer platforms or runtime and orchestration systems in production
- Hands-on experience building agent harnesses or agentic systems: runtime, tool and MCP integration, orchestration, evaluation and guardrails
- Solid software engineering, primarily in Python, with the depth to build production systems and abstractions
- Strong cloud-native background: Kubernetes, containers, Infrastructure as Code, and one major cloud (AWS, Azure or GCP)
- Working grasp of how LLM agents fail and the engineering that makes them reliable: context design, loop and failure detection, verification and evaluation
- Experience treating a platform as a product: contracts, SLOs, adoption metrics and real users
- Professional working English and Spanish
- Nice to have Agent frameworks such as LangGraph, CrewAI, AutoGen etc. and the Model Context Protocol
- Serious production use of AI-assisted development tools (Claude Code, Copilot, Codex)
- LLMOps, agent evaluation, or observability for agentic systems
- Prior work on harnesses that stay stable across model upgrades
- Platform engineering in regulated environments (finance, insurance, public sector)
Apply with one swipe on Sorce. We auto-fill applications and apply on your behalf — no cover letters, no 40-minute forms.
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.