Data scientist (marketing)

Full-Time (Hybrid: minimum two days per week in our London office), PAYE Level: Mid, 2-3 years of relevant experience.

About Fable Data: Fable Data is a global consumer transaction data company. We aggregate anonymised consumer data from financial services businesses, which we then enrich, productise and deliver high value data products to some of the world’s leading retailers, investment managers, technology companies, governments, and advertising firms. Our data provides a near real-time view of the consumer economy offering powerful insights into consumer behaviour, retailer performance and broader macroeconomic trends.

We’re looking for a talented Data Scientist ready to take the next step in their career, someone who thrives on analysing text data and is adept at using AI alongside an expansive machine learning toolkit to build high precision solutions to identify real world entities within billions of lines of text data. With access to one of the most comprehensive, market leading, multi-country consumer transaction datasets available, you will:

• Expand the merchant vocabulary (named entity recognition)
• Build new models and enhance the accuracy of our existing models that power our world class products and the high impact insights produced by our client enablement and commercial teams

Passionate about solving real-world problems through a blend of applied data science, analytical thinking and research. Product driven thinking enables you to systematise your work into reusable and repeatable processes that can be integrated easily into our data platform. Thrive in a fast-paced, collaborative environment that values both analytical rigour and commercial impact.

Contribute to the full ML lifecycle including:

• Model training, evaluation, versioning, deployment, and iterative improvement for a suite of text-based classification models
• Evaluate and validate new data sources for suitability, quality, and bias in ML training pipelines
• Assist in developing and implementing efficient strategies for creating high-quality labelled training datasets, leveraging automation, weak supervision, and active learning techniques
• Design, implement, and maintain rule-based data processing logic leveraging regex and other pattern-matching approaches
• Assist in developing monitoring systems for in-life machine learning models that automatically detect and flag issues
• Work with stakeholders to define and implement new machine learning applications based on transaction data

Experience in SQL and Python in a professional context. Comfortable working with data cleaning, transformation, and basic scripting tasks. Strong attention to detail and a focus on data quality. Experience monitoring and enhancing in-life ML Models (MLOps). Familiarity with classification, time series, and/or natural language processing. Knowledge of or experience working with consumer data, banking data, or stocks and shares. Planning skills to help you prioritise work across multiple projects. This is a unique opportunity to combine technical depth with commercial storytelling and have your work seen by some of the most influential organisations in the world.