Motorway is the UK's largest online car-selling platform. We connect private sellers directly with over 8,000 dealers nationwide. Our online platform helps sellers achieve great prices for their cars while giving dealers fast, reliable access to the stock they want for their dealerships. Founded in 2017, our award-winning, technology-led approach has redefined the experience of selling a car. Motorway is backed by some of the world’s leading technology investors, having raised £143 million in Series C funding.
This is a unique opportunity to join a fast-growing scale-up at a crucial phase of growth and help change an industry for the better.
GenAI Engineering sits within Motorway's Data and AI function, alongside Machine Learning and the Machine Learning Data Platform. We own the AI features that shape how buyers and sellers experience the marketplace, from agentic workflows in the customer journey to LLM-powered tooling for our dealer network.
What we build goes into the hands of real sellers and thousands of verified dealers, usually within weeks. When it works, someone sells their car more easily. When it doesn't, we hear about it. That feedback loop is short, and it shapes how we work. We have a track record of successful deployments and a strong reputation as a result.
We're looking for a Senior GenAI Engineer to build AI applications and agentic workflows for Motorway, taking them from rapid prototype through to production-grade systems that deliver measurable value.
You'll take a problem that isn't fully defined yet and see it through to a reliable system running in production: the prompting, the retrieval, the agent design, the evals, the guardrails, the monitoring. You own that whole stack. There's no separate team who "productionises" your work later.
Most of what makes a GenAI feature good sits around the model rather than in it. The prompting is rarely the hard part. Retrieval quality is, and so is the data feeding it, evaluation you can trust, sensible behaviour when things fail, and getting it live and keeping it there. If that's the work you want to do, we're not precious about the route you took to get here. Some of us came through data science and machine learning and deliberately built up our software engineering; others came the other way.
Own AI features end to end, from an ambiguous problem through to a reliable, observable system that real customers depend on.
Build the retrieval and data foundations these features stand on, and treat retrieval quality as a first-class engineering problem rather than a tuning exercise.
Design evaluation and monitoring approaches that make quality, reliability and safety measurable rather than assumed.
Build AI applications and agents using LLMs, orchestration, APIs and data systems, applying solid production software practices throughout.
Make the calls on models, cost and latency that shape what a feature costs us to run, and document the reasoning so others can follow it.
Prototype quickly, work out what is reusable, and harden the patterns that work into components the rest of the team adopts.
Raise the quality of the work around you through code review, design feedback and the standards you set by example.
Represent your technical decisions directly to product managers and stakeholders, translating model behaviour into business consequence.
You've shipped AI or ML systems that real users depend on, and you've stayed close enough to production to know how they behave when they break.
Strong Python, and SQL you genuinely use. A lot of this work is data work: assembling grounding data, building golden datasets, and digging into why something failed.
You've invested deliberately in your software engineering craft: testing, error handling, CI/CD, observability, and the MLOps practices that keep things healthy once they're live.
You're hands-on across the GenAI stack: LLM APIs, retrieval and vector stores, agents with tool use, and structured outputs, running on cloud infrastructure (we use AWS and GCP).
You have a real point of view on evaluation, most likely because you've been caught out by a system that looked fine and wasn't.
You can reason about how models fail, including hallucination and prompt injection, and what that means for someone who trusts the output.
You know when to reach for a model, when something simpler will do, and when to stop.
You're a constructive reviewer, and colleagues are better off for having worked alongside you.
You're curious about how things actually work, and you follow that curiosity past the first plausible answer.
You're drawn to problems where the right approach isn't obvious yet.
You're honest about the limits of what you know and pull people in early rather than late.
You take responsibility for what you ship, watch how it behaves, and fix it when it breaks.
You get satisfaction from the unglamorous work that makes an AI feature actually good, not just the part that demos well.
You can disagree well, and commit once a decision is made.
We've got your back, in life's biggest moments and the everyday ones too.
Competitive salary
Equity scheme - we all share in Motorway's success
25 days holiday
Flexible working (2 days a week in our London office, with socials and snacks galore)
Private medical insurance via BUPA
Life assurance
Pension scheme (we contribute 5%)
Enhanced family leave (e.g. 26 weeks full pay for maternity or adoption)
24/7 Employee Assistance Programme
EV leasing scheme
Cycle to work scheme
Nursery salary sacrifice scheme
1 paid volunteering day a year
Designs and implements generative AI solutions for customer environments, taking LLM, RAG, and agent-based systems from prototypes to scalable production integrations across cloud platforms and business processes.
Build and operate a secure, reliable AI agents platform, owning its runtime, GenAI gateway, rollout controls, observability, and evaluation infrastructure for healthcare products.
Desenvolve e evolui soluções de IA integradas ao OMS, usando LLMs, prompts e pipelines RAG para automatizar fluxos, otimizar o processamento de pedidos e reduzir trabalho manual operacional.
Build autonomous AI agents for executive workflows, define evaluation systems that measure task quality, and investigate failures to improve reliability before capabilities reach users.
Design and implement production-ready AI agents, prompt-based solutions, and automated workflows for customer-service and payment processes, using Python and Azure Databricks while evaluating quality and reliability.
