Senior AI Engineer

AI Chopping Block; City of Edinburgh; Hybrid; GBP 90,000 - 140,000; Full time

Benefits offered by this job

Time off 27 days + extra days

Health & Wellness benefits

Hybrid work policy

Work-from-anywhere scheme

Space to connect

Job summary

Multiverse is the UK’s leading apprenticeship and AI-first education company, seeking an experienced Production AI Agent Engineer. You will design and own end-to-end agent systems, optimize context management, and drive tool integration across the stack.

You’ll build evaluation frameworks and guide architectural decisions for scalable AI products in a fast-paced environment. You will collaborate with cross-functional teams, review code, and mentor others, using Claude Code to shape

Qualifications

• Shipped AI systems at meaningful scale.
• Experience designing context strategies and retrieval.
• Strong understanding of tool design and system integration.

Responsibilities

• Own and deliver complete agent systems across architecture, implementation and production operation.
• Design and optimize context windows, retrieval pipelines and memory strategies.
• Build evaluation pipelines with metrics for safety, accuracy and latency.
• Design and integrate tools, APIs and data contracts for agents.
• Influence architectural direction with evidence-based proposals.
• Mentor peers through code reviews and pairing.
• Lead with Claude Code as the primary development workflow.

Skills

Context management

Model selection

Cost engineering

Tool integration

Evaluation frameworks

Product thinking

Full-stack delivery

Communication

Tools

MCP (Model Context Protocol)

Claude Code

LLM integration

APIs

Job description

Multiverse is the upskilling platform for AI and Tech adoption.

We have partnered with 1,500+ companies to deliver a new kind of learning that's transforming today’s workforce.

Our upskilling apprenticeships are designed for people of any age and career stage to build critical AI, data, and tech skills. Our learners have driven $2bn+ ROI for their employers, using the skills they’ve learned to improve productivity and measurable performance.

In April 2026, we announced $70 million in strategic funding, led by Schroders Capital, with participation from StepStone Group, Lightspeed Venture Partners and General Catalyst. At an increased valuation of $2.1bn, the round makes us Europe’s first EdTech double unicorn.

But we aren’t stopping there. With a strong operational footprint and 800+ employees, we have ambitious plans to continue scaling. We’re building a world where tech skills unlock people’s potential and output.

Join Multiverse and power our mission to equip the workforce to win in the AI era.

The Role

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. Multiverse is in a uniquely strong position to do that, and getting it right has implications beyond the company: for the UK tech sector and the broader economy.

The AI Transformation team exists to make that real, starting with Multiverse itself. This is not a team that bolts AI onto the edges of the business or ships a handful of internal productivity tools. The mandate is bigger: to rebuild how the company actually works, function by function, and to establish the practices that make Multiverse an AI-first company from the core out.

That work matters twice over. Get it right inside Multiverse and we move faster, serve learners better, and operate at a level few organisations can match. But Multiverse also exists to build the workforce that every other company is reaching for. The way we transform ourselves becomes the standard we set for everyone else. You are not just changing one company, you are building the blueprint others will follow.

The team is one small, focused squad, accountable for outcomes end to end. You work closely with the wider engineering org building Multiverse's customer-facing product, and alongside the teams whose work you are helping to reinvent. The structure is flat and fast. No shared queues, no bureaucratic overhead between having an idea and shipping it.

Whilst we are building something entirely new, Multiverse has an established product, existing infrastructure, and engineering teams in London and Berlin. You need to be as comfortable integrating existing systems and working across team boundaries as you are building new ones from scratch.

What You Will Do

Own and deliver complete agent systems. You take a product problem and build the agent system that solves it. Architecture, implementation, evaluation, and production operation. You are responsible for the system working, not just for your code compiling.

Design context and retrieval strategies. What goes into the context window and what stays out is the most consequential design decision in an AI system. You design retrieval pipelines, conversation memory, summarisation strategies, and the chunking logic that makes context useful rather than noisy. You understand the cost and quality trade-offs at every layer.

Build evaluation frameworks. You define and implement the metrics that tell the team whether its AI systems are doing what they should. Accuracy, safety, helpfulness, domain-specific quality, latency. You build automated eval pipelines and human-in-the-loop review processes. You treat evaluation as an engineering discipline, not an afterthought.

Design tool integrations. Agents are only as capable as the systems they can reach. You design and build the tool layer: MCPs, APIs, data contracts, and the error handling that makes tool use reliable. You work closely with the wider engineering org building Multiverse's customer-facing product, whose systems your agents need to interact with.

Influence technical direction. You have opinions about how things should be built, and you back them up with evidence. You contribute to architectural decisions, push back when the team is heading in the wrong direction, and propose better approaches. You are not a team lead, but your technical judgement shapes what gets built and how.

Raise the bar through code review and pairing. You review code with rigour and give feedback that makes the team better. You pair with less experienced engineers on hard problems. You set a standard for what production-quality AI engineering looks like.

Use Claude Code as your primary development workflow. Claude Code is how this team builds. You set context, define constraints, review output critically, and augment the tool with skills and domain context. You are fluent in AI-assisted development and can mentor others in doing it well.

What We Are Looking For

Production AI Agent Engineering

You have shipped AI systems that serve real users at meaningful scale. You understand the engineering challenges that make agent systems a different discipline from conventional software:

• Context management. Designing what enters the context window and what stays out. Retrieval strategies, chunking approaches, conversation memory, summarisation. You know how context quality drives output quality and cost, and you have made these trade-offs in production.

• Model selection and routing. Choosing the right model for a task based on capability, latency, cost, and reliability. You have worked with multiple models and understand when a smaller, faster model is the right call.

• Cost engineering. Token economics, caching, prompt optimisation, batching. You know the difference between a prototype that works and a production system that works at a cost the business can sustain.

• Tool use and agent augmentation. Designing the tool surfaces that agents use to interact with external systems. Writing tool descriptions that models use reliably, handling failures gracefully, building integration layers that are composable rather than brittle.

• Evaluation. Building frameworks for assessing AI output quality: accuracy, safety, helpfulness, domain-specific criteria. You ship with eval, not after it.

Product Thinking

You do not wait for a spec. You understand the problem, figure out what needs to exist, and build it. On a small squad there is no gap between product thinking and engineering. You talk to users, understand their workflows, and identify the highest-value intervention.

This does not require product management experience. It requires the instinct to ask “what problem are we solving and for whom?” before “what framework should we use?”

Full-Stack Delivery

You work across the stack: LLM integration, backend services, data pipelines, and enough frontend to ship end to end. Agent systems do not fit neatly into service boundaries, and your ability to work across all of them is a practical requirement.

Communication

You explain technical decisions clearly to both engineers and the product and design people you work with day to day. You document your designs, write pull requests that tell a story, and give direct feedback without being abrasive.

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

• Published thinking or external contributions in AI engineering (talks, writing, open source)

• Experience with multi-agent coordination: task decomposition, handoff, shared state

• Practical experience with MCP (Model Context Protocol) or equivalent agent integration standards

What We Are Not Looking For

• Pure ML researchers without production engineering experience. We build products, not papers

• Narrow specialists. If you only do infrastructure, or only do model training, or only do frontend, this team needs broader range

• Engineers who need a detailed spec and a sprint plan before starting. We ship fast and iterate

• Candidates whose AI experience stops at wrapping LLM APIs. We need depth in context strategy, evaluation, tool design, and the systems engineering underneath

• Engineers who optimise for technical elegance over user outcomes. The architecture serves the product

Benefits

• Time off - 27 days holiday, plus 5 additional days off: 1 life event day, 2 volunteer days, 2 company-wide wellbeing days (M-Powered Weekend) and 8 bank holidays per year

• Health & Wellness- private medical Insurance with Bupa, a medical cashback scheme, life insurance, gym membership & wellness resources through Wellhub and access to Spill - all in one mental health support

• Hybrid work offering - for most roles we collaborate in the office three days per week with the exception of Coaches and Instructors who collaborate in the office once a month

• Work-from-anywhere scheme - you'll have the opportunity to work from anywhere, up to 10 days per year

• Space to connect: Beyond the desk, we make time for weekly catch-ups, seasonal celebrations, and have a kitchen that’s always stocked!

Our Commitment to Diversity, Equity and Inclusion

We’re an equal opportunities employer. And proud of it. Every applicant and employee is afforded the same opportunities regardless of race, colour, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender, gender identity or expression, or veteran status. This will never change. Read our Equality, Diversity & Inclusion policy here.

Our Commitment to Safeguarding

Multiverse is committed to safeguarding and promoting the welfare of our learners. We expect all employees to share this commitment and adhere to our Safeguarding Policy, our Prevent Policy and all other Multiverse company policies. Successful applicants will be required to undertake at least a Basic check via the Disclosure Barring Service (DBS).

For roles that will involve a Regulated Activity, successful applicants must also undergo an Enhanced DBS check, including a Children’s Barred List check and a Prohibition Order check. Roles involving Regulated Activity may interact with vulnerable groups, therefore are exempt from the Rehabilitation of Offenders Act 1974 meaning applicants are required to declare any convictions, cautions, reprimands, and final warnings.

Providing false information is an offence and could result in the application being rejected or summary dismissal if the applicant has been selected, and possible referral to the police and the DBS.