Yonder
Data Scientist
Get notified of more jobs like this · No spam, ever
How your CV stacks up
Upload your CV to see how well it fits this job role
?%
What’s Yonder?
“It's as if Time Out, Amex and Monzo had a baby” - Will T, Yonder Member
Yonder is a credit or debit rewards card that is designed to be fair, flexible, and actually enjoyable to use. No confusing terms. No dusty points system. Just rewards that actually feel rewarding - from bao to beers to a boarding pass.
If you want to work somewhere that's at a genuine inflection point, and on a product that our members really love, come Yonder with us.
Sounds cool. What’s my part in this?
We’re looking for a Senior Data Scientist to build the predictive models that sit behind some of Yonder’s most important decisions.
You’ll work across credit acquisition, portfolio management, collections, fraud and rewards, using our data to improve how we select customers, manage risk, allocate credit, intervene when things go wrong and personalise the Yonder experience.
This is a senior, hands-on individual contributor role reporting directly to the CRO, who owns Risk and Analytics at Yonder. You won’t be sitting in a central data science team waiting for problems to arrive. You’ll work directly with the people making the decisions, understand the commercial problem, build the model and help turn it into something that actually runs in production.
That means everything from feature engineering and model development through to validation, implementation, monitoring and explaining what the model is telling us. We care far more about models that improve decisions than models that are technically interesting but never make it into the product.
Yonder moves quickly. We have a growing customer base, increasingly rich behavioural data and a lot of decisions that can be made materially better through predictive modelling. We’re looking for someone excited by the opportunity to build that capability rather than inherit a finished machine learning stack.
What you’ll do
- Building our acquisition models. You’ll develop and continuously improve models that help us understand applicant risk, affordability and expected customer value. You’ll use bureau, application and alternative data sources to make better onboarding decisions.
- Building behavioural models for portfolio management. You’ll predict how existing members are likely to behave to help our credit strategy team to use those predictions to improve credit profitability.
- Improving collections strategy through prediction. You’ll build models that help us identify how members will respond to different interventions and allow our Collections team to focus their efforts where it has the greatest impact.
- Building better fraud models. You’ll work with our Fraud and Financial Crime team to identify suspicious behaviour earlier, improve fraud detection and reduce unnecessary friction for genuine members.
- Using machine learning to improve rewards. You’ll work closely with our AI engineers to leverage member data to help surface the right reward at the right time.
- Owning the modelling lifecycle. You’ll take models from an initial hypothesis through data exploration, feature engineering, training, validation, implementation and ongoing performance monitoring. Getting these models into production is also part of the role, not an engineer’s.
- Identifying the data that makes our models better. You’ll constantly look for additional internal signals, bureau attributes and external data sources that could materially improve prediction. We expect you to challenge whether the data we have today is enough.
- Building strong model monitoring and validation. You’ll make sure we understand how models perform after deployment. Does it discern better? Is it well calibrated? Are we seeing drift?
- Explaining complex modelling simply. You’ll translate complex outputs to the non-technical. Being able to clearly explain why a model has been built the way it has been to the CRO and Committee members is important.
- Helping shape how Data Science works at Yonder. This is still a capability we are building. You’ll help establish the modelling standards, tooling, experimentation practices and development approach that allow us to move quickly without losing rigour.
Reasons to use Rodeo
Start with a chat, not a search bar
Grad scheme, placement, apprenticeship? Not sure what you want yet — that's fine. Your agent talks it through with you and turns "I have no idea" into a shortlist.
It searches the market for you
Every day your agent scans the market matching roles against what actually matters to you, not just keywords on a CV.
Only hits
No noise. No "maybe this fits." Just roles with a clear explanation of why they're right — and where to focus when applying.
You're a great fit if you
- ✅ You’ve built predictive models that drove great decisions. You can point to models you developed that made it into production and materially improved risk, revenue, customer experience or operational outcomes.
- ✅ You love being hands on. Python is second nature to you, you’re highly capable in SQL, and you’re comfortable working directly with large, messy datasets without needing an analyst or data engineer between you and the problem.
- ✅ You understand predictive modelling deeply. You’re comfortable with approaches ranging from logistic regression and gradient boosting. Choosing the right algorithm given the problem statement is just as important as being able the model.
- ✅ You understand how to evaluate models properly. You think beyond headline accuracy metrics and care about calibration, stability and drift. These are production models and impacts live decisions.
- ✅ You think commercially as well as statistically. You understand that the best model isn’t necessarily the one with the highest Gini. You care about the actual outcomes it could effect.
- ✅ You can operate with ambiguity. Sometimes the question will be clearly defined. Other times you’ll be given a problem statement that needs to be solved. You’re comfortable looking for a solution and also reaching out for help where needed.
- ✅ You’re comfortable owning your work end to end. You want responsibility for the model’s outcomes, including implementation, monitoring and adjustments.
- ✅ You communicate technical work extremely well. You can explain what a model is doing, where it works, where it doesn’t and generally how it should be used without using technical language.
- ✅ Experience in consumer lending, credit cards, fraud or fintech is valuable. Particularly if you have worked with bureau data, behavioural credit models, collections models, transaction data, fraud detection or customer recommendation systems.
- ✅ You have a builder's instinct. You like getting models into the hands of decision makers, measuring whether they worked and improving them quickly rather than spending months trying to build something theoretically perfect.
Get help with your application
Your very own career expert that helps elevate your application to the next level.
You won’t be a great fit if you
- 👎 You mainly want to do exploratory analysis. Analysis will be part of the job, but the expectation is that it ultimately results in models, decisions and measurable outcomes.
- 👎 You want to hand models over after development. You’ll be expected to care about implementation, monitoring and performance once a model is live.
- 👎 You optimise for modelling sophistication over business impact. We’re very happy using a simple model when it makes the best decision. Complexity should be based on need.
- 👎 You need perfectly structured datasets and fully defined problems. Yonder is growing quickly. Some of the most valuable work will involve creating structure where it doesn’t yet exist.
- 👎 You want distance from commercial decisions. You’ll work directly with the CRO and teams across Credit Risk, Fraud, Collections, Finance, Product and Marketing. We expect you to have a view on what the analysis means and what we should do with it.
- 👎 You want established modelling playbooks to follow. We have a strong data foundation, but there is still a lot to build. You’ll have significant influence over how predictive modelling develops at Yonder.
- 👎 You want a fully remote role. We're office-first. You'll be in the office at least three days a week because we think the best problem solving happens when Data Science is close to the teams using its work.
What’s it like working at Yonder?
- 🏢 We’re office-first, remote-friendly We’re based in our London Shoreditch office, complete with dog visitors and plenty of comfortable space to do your best work. We ask you to come into the office at least 3 days a week, with everyone coming in on Mondays.
- 🤍 We take a values-led approach Our principles are incredibly important to us, so we recommend you check them out here: Our DNA
What’s in it for me?
Depending on your skill set and what you can bring from day one, you’ll be looking at:
- 💰 £ £94,589 - £111,502 annual salary depending on experience
- 📈 £ £80,479 - £105,003 stock options
- Plus
- ✈️ 35 holidays (27 days annual leave + 8 days public leave)
- 🐶 Team-building offsite
- ❤️🩹 Private healthcare with Bupa
- 🐣 Up to 12 weeks enhanced parental leave after being with Yonder for 1 year
- 🧠 Learning & training allowance (£750/year) that you can use on books, courses, etc
- 🍳 Regular team breakfasts and lunch
- ⛳️ Regular team events like Mini-golf, Escape Room, Cocktail making
- 🚴 Cycle-to-work scheme
- ☕️ Fresh pour-over coffee made by our very own CEO, Tim (London only)
What’s the interview process like?
We take the candidate experience really seriously, so we’ve made the process as transparent as possible. We also promise to be super responsive, and will never leave you wondering where you stand for weeks on end.
Here’s how it works:
- Stage 1: Intro call (45 mins): You will have an initial Zoom call with [Hiring Manager] to find out more about you and to tell you more about us.
- Stage 2:
“It took my CV and asked me questions relevant to understanding what kind of jobs to suggest for me. Suggestions were almost perfect. Jobs were exactly what I’ve been looking for.”
Jessica, London