Responsibilities

• design and maintain the data pipelines and ML-ready tables that power feature generation, model training, inference and impact measurement – with clear ownership of data quality, reproducibility and training-serving consistency.
• help the CRM tribe find the biggest opportunities for growth and personalisation across the customer lifecycle — onboarding, engagement, activation, retention.
• own predictive/uplift or recommendation models that decide what to recommend and who to talk to, powering personalised CRM journeys end-to-end.
• model customer behaviour, product-usage patterns and CRM engagement (email/push clicks) to identify who's likely to churn, who's ready for their next product, and measure how much value a campaign actually created.
• partner closely with Data Analysts, Data Engineers and CRM Campaign Managers, translating models into levers they can use.
• average day will include building or maintaining production models, running experiments, evaluating new ideas, and communicating what models can (and cannot) tell us about how and why CRM drives growth.
• Have direct impact — you will closely partner with the CRM and CRM Analytics team to help ship models that reach every Wise customer through CRM campaigns and personalisation surfaces.
• Own the problems worth solving — you will not just be handed a backlog. You will form strong opinions on where DS creates the most value and convince the people around you to do what is necessary to help.
• Work autonomously - ... freedom to make your own calls.
• Be part of a diverse team - You will work in a team of Data scientists, Analysts, Engineers and CRM Managers.

Requirements

• You have expert knowledge of Python and are able to make and justify design decisions in your code (packaging, testing, config-driven pipelines).
• You have expert knowledge of SQL and can write, debug and optimise complex queries against a warehouse.
• You have hands-on experience shipping and operating predictive ML models or recommendation systems in production (classification / propensity / churn / next-best-action) — end-to-end from feature pipelines (SQL or feature store) and ML-ready training / scoring datasets, through model registries, batch or real-time inference, and monitoring — with sound data-modelling practices and data-quality checks along the way. You know when to reach for gradient boosting, neural networks, linear models, or a blend.
• You have designed and analysed A/B tests end-to-end — power calculations, choosing the right unit of randomisation, guardrail metrics, interpreting inconclusive results.
• You have a proficient understanding of statistics — sample size, variance, multiple-comparisons, Bayesian reasoning.
• You know how to measure the incremental impact of ML models and CRM interventions end-to-end — defining success and guardrail metrics, designing and powering A/B tests, choosing the right unit of randomisation, quantifying uncertainty, accounting for multiple comparisons, and interpreting inconclusive results to inform business decisions.
• You take end-to-end ownership with a structured, data-driven approach — cutting through vagueness to frame the business problem, prioritising the value you can add, and defining precisely where and how a model fits into the stack.
• You communicate effectively with any audience — translating model choices, uncertainty and trade-offs into decisions non-DS teams (Product, Marketing, CRM Ops, Compliance) can act on, and visualising data clearly along the way.
• You operate beyond your direct team — shaping design decisions on the product or engineering side, mentoring other DS and analysts, spawning and running cross-functional projects through to delivery, educating the wider company on what DS can do, and keeping current with developments in your areas (recommenders, uplift, causal inference).

Preferred

• You have hands-on experience with sequential and/or deep-learning recommender models — two-tower architectures, transformer-based rankers, or sequence embeddings.
• You have experience with uplift modelling and conditional average treatment effects for personalisation.
• You are familiar with dbt, Snowflake, AWS (S3, SageMaker), Airflow, and Trino / data lake stacks.
• You have prior CRM / marketing / e-commerce ML experience — customer lifecycle segmentation, campaign attribution, contact strategy optimisation.

Benefits

• Flexible working - whether it’s working from home, school plays or life admin we get that flexibility is essential and you’re trusted to do the right thing and be responsible
• Stock options in a profitable company
• Relocation support
• Generous parental leave
• Pension scheme
• Paid sabbatical
• Loads of development opportunities
• Sports and wellbeing compensation
• A fun work environment with social activities and events
• The opportunity to work with super smart, curious people