Role Overview

We’re looking for a Data Scientist to join our Technology Solutions squad. You’ll develop the models behind our custom-built solutions for sports events and properties, contributing to a growing portfolio of broadcast, digital and fan-facing products delivered via B2C/B2B applications and APIs. You’ll sit within a cross-functional squad and work closely with colleagues across the business, so you’ll need to be comfortable collaborating across disciplines and different technologies.

What You’ll Do

• Modelling & Analysis: Develop, train and evaluate models using statistical and machine learning techniques, with a focus on probabilistic approaches. Contribute across the modelling lifecycle from feature engineering and training through to validation and deployment.
• Data Work: Query, clean and explore datasets using Python and SQL to surface patterns and support model development.
• Data Pipelines: Help build and maintain the pipelines your models depend on, ingesting and validating new and often messy sports data sources.
• AI-Assisted Development: Leverage AI tools to accelerate and improve your day-to-day workflow.
• Event Support: Our solutions are often built around specific sporting events, meaning fixed deadlines and go-live support requirements that you will help to provide.
• Quality & Rigour: Apply good model development discipline through version control, testing and documentation.

What You’ll Bring

• Passion for Sport: You follow sport closely and understand the context of the data and audiences we build for. Comfortable with sport-driven modelling decisions.
• Machine Learning & Statistics: Solid grounding in machine learning, supervised and unsupervised methods, and classical statistical techniques. Comfortable working with probabilistic models, uncertainty estimation and Bayesian inference.
• Model Development: Understanding of the full model training pipeline, including data preparation, feature selection, model selection and model validation.
• Experience: Hands-on experience building and evaluating models in a data science or quantitative context.
• Python & SQL: Comfortable using Python and SQL for data exploration, feature development and modelling workflows.
• Interest in Data Engineering: An appetite for the engineering side of the work — you want to understand and help own the pipeline that feeds your model, rather than hand that problem to someone else.
• Client-Centricity & Communication: You keep the client in mind throughout, and can present findings clearly to both technical and non-technical audiences. You’re comfortable explaining your work directly to client stakeholders and translating what they need into modelling decisions.