Data Scientist

Data Scientist | 📍London or Manchester – Hybrid (1–2 office days per week) | 💰Competitive Salary + Benefits

About the Role

We’re looking for a Data Scientist to join Moonpig, working hybrid from London or Manchester. You’ll build, evaluate and help productionise machine learning solutions that solve real customer and commercial problems across recommendations, personalisation, customer modelling and predictive modelling.

This is a hands-on applied Data Science role where you’ll work closely with Product, Engineering, MLOps, Commercial and Marketing. You’ll turn clearly defined problems into practical ML solutions, evaluate whether they’re working and help bring them successfully into production.

You’ll have the independence to make sound decisions within your problem space, while being part of a collaborative team that values high-quality, reproducible code and thoughtful experimentation. You’ll also use modern AI-assisted development tools responsibly to improve the speed and quality of delivery.

• Develop and evaluate machine learning models across recommendations, personalisation, customer and predictive modelling.
• Explore data, engineer useful features and compare modelling approaches, choosing solutions that fit the problem rather than adding unnecessary complexity.
• Partner with Product, Commercial, Marketing and other stakeholders to understand problems, clarify requirements and translate them into practical Data Science approaches.
• Apply appropriate offline model evaluation, investigate model behaviour and clearly communicate performance, limitations and trade-offs.
• Contribute to the design and analysis of A/B tests and other experiments, connecting model performance with customer behaviour and business outcomes.
• Develop solutions with production use in mind, partnering with Engineering and MLOps to integrate models into ML pipelines and support deployment, monitoring and ongoing improvement.
• Write tested, modular and maintainable Python and SQL, contributing to shared codebases and reproducible workflows using established software-development and version-control practices.
• Monitor deployed solutions and investigate model performance, data quality and unexpected behaviour, contributing improvements where needed.
• Use AI-assisted tooling across coding, analysis, exploration, experimentation and documentation, critically validating outputs to maintain quality.
• Take part in code and analytical reviews, share knowledge and contribute to reusable tooling, documentation and improvements to Data Science ways of working.

Python and SQL for modelling, analysis and production Data Science.

AWS for cloud-based data and machine learning.

Git and shared codebases supporting version control and collaborative development.

ML pipelines supporting integration, deployment, monitoring and iteration.

A/B testing and experimentation to connect technical model performance with customer and business outcomes.

AI-assisted development tools used across coding, analysis, experimentation and documentation.

dbt is part of our wider analytics engineering tooling.

You’ll be a reliable, independent contributor within a defined problem space. That means understanding the relevant data, selecting an appropriate approach, building and evaluating a solution, communicating what you’ve learned clearly and working with others to put that work into practice.

Success will come through consistently delivering high-quality modelling and analytical work, making sensible technical choices and building maintainable, reproducible solutions that work effectively within production ML workflows.

You’ll use evaluation and experimentation to understand whether solutions are making a difference, while collaborating across Data Science, Product, Engineering, MLOps and our business teams. You’ll also help strengthen the wider Data Science team through high-quality code, constructive reviews, knowledge sharing and reusable tools.

Following an initial recruiter screening, the expected process includes a Hiring Manager interview, Technical Screening, Technical Interview follow-up and Final Round.

The exact structure is still being confirmed, and we’ll keep candidates informed of any changes throughout the process.

Experience developing machine learning or advanced analytical solutions in a Data Science, Machine Learning or Advanced Analytics role.

Strong practical understanding of supervised machine learning, feature engineering, validation, overfitting and model evaluation, backed by real-world modelling experience.

Strong Python and SQL skills, with experience applying both to real-world data and modelling problems.

Ability to translate defined customer or business problems into appropriate analytical or machine learning approaches.

Experience selecting and applying model evaluation metrics and validation approaches, with an understanding of their strengths and limitations.

Experience designing or analysing A/B tests or other controlled experiments, including selecting success metrics and interpreting results.

Experience with Git or similar version-control tools and contributing clear, modular and maintainable code to shared codebases.

Understanding of testing, reproducibility and good software-development practices.

Experience contributing to production machine learning workflows, including an understanding of deployment, monitoring, data quality and the wider model lifecycle.

Ability to explain assumptions, methods, results and technical trade-offs clearly to both technical and non-technical audiences.

Comfortable independently delivering defined modelling or analytical work and knowing when to seek input on unfamiliar or more complex problems.

Comfortable using AI-assisted development tools for coding, analysis or experimentation, with the judgement to critically evaluate their outputs.

Awareness of data quality, privacy, fairness, security and customer-experience considerations when developing data-driven products and solutions.

Experience in B2C e-commerce, retail or a high-volume digital environment would be useful, but isn’t essential.

Experience with recommendation or personalisation systems would be beneficial.

Exposure to customer modelling approaches such as propensity, uplift or customer lifetime value modelling would be beneficial.

Experience applying LLMs, embeddings or other generative AI capabilities to practical product, analytical or Data Science problems would be useful.

Experience with cloud-based data or machine learning platforms, particularly AWS, would be beneficial.

Familiarity with analytics engineering tooling such as dbt would be useful.

A degree in Statistics, Mathematics, Economics, Computer Science or another relevant quantitative discipline can be helpful, but equivalent practical experience is equally welcome.