Senior AI Engineer, Edinburgh

The Opportunity

Multiverse is the UK’s largest apprenticeship provider and its first EdTech unicorn. The current state of AI presents a huge opportunity to reshape the future of education and workforce development — and Multiverse is in a uniquely strong position to do that. Getting it right has implications beyond the company: for the UK tech sector and the broader economy.

The Scotland hub exists to make that real. A new engineering team with the mandate to build AI-native products, help modernise the existing platform, and set the practices that make Multiverse an AI-first company. Multiverse has built an environment where AI-native ways of working collapse the old boundaries, so one person can own the whole arc from idea to live product.

As an AI Engineer at P6, you’re a specialist and a core builder. You’re the go-to person for your product domain — someone who can take a hard problem, make the right design calls within it, and ship something that works for real users. You won’t be waiting for the work to be broken down for you. You’ll work in a small, focused squad led by a Principal engineer, with full ownership of your slice of the system from design through to production.

What You’ll Do

• Own and deliver agent features end-to-end. Take a product problem — a coaching workflow, a content pipeline, a retrieval system — and build the agent feature that solves it. Architecture within your domain, implementation, evaluation, and production operation. You are responsible for it working, not just for your code compiling.
• Design context and retrieval strategies. Decide what goes into the context window and what stays out. Build retrieval pipelines, conversation memory, and summarisation logic that makes context useful rather than noisy. Understand the cost and quality trade-offs at every layer.
• Build and maintain evaluation frameworks. Define the metrics that tell the team whether its AI systems are doing what they should. Build automated eval pipelines and human-in-the-loop review processes. Treat evaluation as an engineering discipline, not an afterthought.
• Design tool integrations. Agents are only as capable as the systems they can reach. Build the tool layer: MCPs, APIs, data contracts, and the error handling that makes tool use reliable across the systems your agents interact with.
• Shape technical direction within your domain. You have strong opinions about how things should be built and you back them up. Contribute to design reviews, push back when the approach is wrong, and propose better paths. Your technical judgement shapes what gets built and how within your squad.
• Raise the bar through review and pairing. Review code with rigour and give feedback that makes engineers better. Pair with less experienced colleagues on hard problems and help set the standard for production-quality AI engineering on the team.

What We’re Looking ForProduction AI Engineering

You’ve shipped AI-powered features to real users. You understand what separates a prototype from a production system: context quality, model selection trade-offs, token economics, reliable tool use, and evaluation that runs before you ship. You don’t need multi-agent architecture at this level, but you build the systems that sit inside one.

Depth in Your Domain

You’re a subject matter expert in at least one area of the AI engineering stack — retrieval, context management, evaluation, tool design, or backend systems that support agents. You can demystify that area for the team and make better decisions within it than most.

Full-Stack Delivery

You work across the stack — LLM integration, backend services, data pipelines, and enough frontend to ship end-to-end. You build with Claude Code daily, set context before generating, and review output critically. AI-native development is how you work, not a shortcut you reach for occasionally.

Product Instinct

You ask “what problem are we solving and for whom?” before “what framework should we use?” You talk to users, understand their workflows, and make calls about what’s worth building without waiting for a spec.

What Would Set You Apart

• Experience building AI systems in EdTech, regulated content, or domains where output quality has compliance or accreditation implications
• Background as a founding or early-stage engineer at a startup
• Experience with multi-agent coordination: task decomposition, handoff, and shared state
• Practical experience with MCP (Model Context Protocol) or equivalent agent integration standards
• Published thinking or external contributions in AI engineering — talks, writing, open source