Quality Engineer

Quality Engineer | £750/day Inside IR35 | 6-month initial contract | London Industry: Technology Location: London Job Type: Contract - 6-month initial This programme delivers a production-grade enterprise agentic AI platform, with MCP acting as the extensibility layer. Responsibilities Design and implement evaluation frameworks for MCP tools and Skills, covering correctness, safety, reliability, and regression performance. Build automated testing pipelines for agentic behaviours, including multi-step workflows and tool-calling interactions.

Identify and mitigate AI failure modes such as hallucination, incorrect tool usage, invalid inputs, and latency amplification. Produce and maintain formal testing evidence required for internal release approvals (build and operate gates). Collaborate with ML Engineers to integrate test hooks and observability directly into MCP servers and Skills.

Contribute to defining long-term quality assurance standards, testing practices, and governance models for a federated MCP ecosystem. Skills & Experience Required Strong Python experience for building test frameworks, automation tooling, and validation pipelines. Experience with LLM evaluation approaches, RAG evaluation frameworks, or custom model assessment systems.

Hands-on experience in test automation across unit, integration, and regression testing. Understanding of agentic AI systems and their failure modes, particularly in non-deterministic environments. Ability to assess systems against governance, security, and audit/compliance requirements.

Experience working within regulated or highly controlled engineering environments.