Applied AI Analyst (Mid and Senior Level) - Transaction Enrichment

Our Story Hello there. We’re Zopa. We started our journey back in 2005, building the first ever peer-to-peer lending company. Fast forward to 2020 and we launched Zopa Bank. A bank that listens to what our customers don’t like about finance and does the opposite. We’re redefining what it feels like to work in finance. Our vision for a new era of banking puts people front and centre — we’ve built a business that empowers everyone to aim high, every day, to move finance forward. Find out more about our fantastic offerings at Zopa.com! We’re incredibly proud of our achievements and none of it would be possible without the amazing team here. It’s not just industry awards we’re winning, we’ve also been named in the top three UK’s Most Loved Workplaces. If you embrace unconventional challenges, are unafraid to think differently and are driven to make an outsized impact, you’ll thrive here at Zopa, so join us, and make it count. Want to see us in action? Follow us on Instagram @zopalife The role Every transaction made by a Zopa customer passes through our transaction enrichment pipeline. We tag the merchant and assign a spending category, so customers can understand their money better and the rest of the bank can build on reliable data. This role is about making that classification measurably better. As an Applied AI Analyst, you will design and tune the prompts that do the categorising, work out where and why they fall short, and help shape what we build next. At heart this is an analytical role, but one where product thinking matters as much as technical ability. You will move quickly, make pragmatic trade-offs, and care about what ships and what lands. You will work closely with the Product and Engineering team to make sure your findings influence the roadmap. You will be comfortable using LLM APIs directly and building with AI coding tools, and you will need the underlying understanding to use them well. About the Transaction Enrichment Team The team sits at the heart of how Zopa turns data into intelligent products. We build and evolve the capabilities that transform raw transaction data into reliable, structured intelligence, so customers can understand their money better, and the rest of the business can build confidently on top of it. A core focus is transaction enrichment: tagging merchants, assigning categories, ensuring data quality, and making sure everything integrates cleanly into our broader data models and customer-facing products. As we move toward more AI-first ways of working, the tribe is also rethinking how transaction data should be accessed and used — not just by analysts and product teams, but by AI systems and LLM-powered tools. The work is foundational and long-term in nature: improving systems, shaping strategy, and building capabilities that compound in value over time. A day in the life • Design, test and refine the prompts that tag merchants and assign spending categories, and work out how transaction data should be prepared and passed to the model. • Analyse enrichment results using Python and SQL to understand where and why enrichment falls short and making evidenced recommendations for improvements. • Surface opportunities the team would not otherwise: spot patterns in the data that point to a product gap, a recurring failure mode that suggests a systemic fix. • Design and run evaluations, including LLM-as-judge approaches, to analyse whether a change improves things or introduces new problems. • Build quick, scrappy prototypes and test harnesses to test a hypothesis with real data before any engineering resource is committed. • Shape the future of the product, for example how customers could re-categorise transactions, and how enriched data can support a wider range of Zopa products and AI-powered tools. About you We hire for capability rather than years, though this is typically 1-3+ years in an analytical or technical role. You’ll likely bring many of the following: • You think in experiments: you form a hypothesis before looking at the data, design something that could prove you wrong, and interpret results with appropriate scepticism. • You are curious about messy, imperfect systems: you want to understand why something breaks, and you brainstorm ideas to improve it. • You can describe what you want and build it together with an LLM, to explore data, clean outputs, call APIs and prototype ideas. • You have some experience with LLM APIs: calling them, prompting them, structuring their outputs, and understanding why they sometimes give you nonsense. • You keep the product goal in view: you care about what changes for customers and the business. • You are comfortable owning ambiguous problems: you can scope the work, prioritise what to test first, and know when good enough is good enough. • You are a clear communicator who can explain what you built, what you found and what it means to engineers, PMs and stakeholders who don’t care about the implementation details. Nice to have None of this is required, and we welcome smart people who have not done it yet. If you have experience of any of the following, we would love to hear about it: • Evaluating LLM outputs, including LLM-as-judge approaches, ground truth sets or evals. • Prompt engineering and prompt tuning, and comparing LLMs on accuracy, cost and coverage. • Classification, categorisation, labelling or entity-matching problems. • Banking, fintech, payments or another fast-paced, data-led environment. • Side projects or things you have built with LLMs simply because you wanted to. What this role is not: • This is not a research role. We value what ships and what lands over what looks good in a notebook. • This is not a production engineering role. Engineering owns production code and deployment. You may occasionally contribute where it makes sense, but that is not your primary responsibility. • This is not a reporting role. Your value lies in the recommendations you make and w