AI QA Engineer

Job title: AI QA Engineer Type: Full-time Contract (until September) Location: UK-based applicants only. You must live within 30 miles of one of our five UK offices: Edinburgh, Newcastle, Leeds, Manchester or London Working pattern: Flexible (remote, hybrid, or in-office) The role We're looking for an AI QA Engineer to work across product quality, AI behaviour evaluation, and defect investigation. This is not a traditional QA role focused purely on scripted testing or automation. A key part of the role involves reviewing AI agent conversations and assessing whether responses were accurate, appropriate, and aligned to the expected customer experience. Because AI behaviour is not always deterministic, strong judgement, attention to detail, and communication skills are essential. You'll support release testing and regression prevention across the platform, carrying out structured manual QA across key journeys during release cycles to improve release confidence and reduce production issues. Outside release periods, you'll help improve automated coverage, refine QA processes, and investigate product or AI-related issues raised internally. You'll also investigate production and customer-reported issues, identifying whether problems are expected behaviour, configuration issues, or genuine defects before escalating clear issues to engineering teams. Required experience Strong manual QA experience Experience building or maintaining automated test coverage Solid, hands-on experience working within an agentic software delivery lifecycle. You should be comfortable with the tooling, workflows, and quality considerations specific to agentic systems, not just AI products in general Strong written English and communication skills The ability to assess nuanced conversations and provide clear feedback Confidence investigating issues independently and making sound judgement calls The ability to think on your feet and reason about unfamiliar problem domains. We'll ask you to do this in the interview, so come ready to engage with specifics rather than general examples Desirable experience Experience working with AI tooling, agent frameworks, or conversational AI platforms is a plus What you'll get A varied role combining QA, AI evaluation, automation improvement, and product investigation rather than repetitive manual testing. The opportunity to work on AI-driven products where quality depends on judgement and context, not just predefined test cases. Real ownership over how quality processes evolve as AI capabilities continue to grow. Exposure to modern AI and agentic systems in a production environment. The chance to work closely with Product, Engineering, and AI teams on complex problems where there often isn't a single correct answer. About hedgehog lab We're a digital product consultancy with 20 years of experience, helping organisations solve complex problems through technology, design, and delivery. Ready to apply? Submit your application below. Please note: we're unable to offer visa sponsorship for this role.