Data Scientist, Starling Bank

• We are looking for talented data professionals at all levels to join the team. We value people being engaged and caring about customers, caring about the code they write and the contribution they make to Starling
• People with a broad ability to apply themselves to a multitude of problems and challenges, who can work across teams do great things here at Starling, to continue changing banking for good
• Industry experience in data science or machine learning models, ideally in AML, financial crime, or a related domain
• Experience working with large-scale, high-dimensional, and heavily imbalanced datasets
• Excellent skills in Python and SQL
• Solid understanding of classification algorithms such as gradient boosting decision trees, including pros and cons of different model architectures
• Strong feature engineering skills and experience in transforming raw data into useful model inputs
• Effective communication skills and able to explain complex findings clearly to both technical and non-technical stakeholders
• Demonstrable experience deploying machine learning solutions in a production environment, and familiarity with version controls systems (e.g. Git)

Desirable

• Our Data teams are excited about the value of data within the business, powers our product decisions to improve things for our customers and enhance effective and agile decision making, regardless of what their primary tech stack may be
• We value autonomy - you’ll be trusted to manage your own projects, drive modelling initiatives, and take ideas from concept to production
• You’ll be encouraged to propose new approaches and explore creative ways to detect and prevent fraud
• We debate and critique our ideas in a healthy, supportive team
• You’ll have the chance to shape both models and how we think about fraud detection as a wider team
• You will be part of a team that builds, evaluates and deploys machine learning models to improve and automate decision making
• Collaborate with technical and non-technical teams to understand problems, explore data, and develop effective fraud prevention tools and solutions
• Design and maintain robust feature engineering pipelines for modelling, working closely with analytics engineering teams
• Contribute to the development of end-to-end machine learning workflows and help embed models into production systems
• Analyse transaction and behavioural data to identify trends, anomalies, and AML patterns

• Interviewing is a two way process and we want you to have the time and opportunity to get to know us, as much as we are getting to know you! Our interviews are conversational and we want to get the best from you, so come with questions and be curious. In general you can expect the below, following a chat with one of our Talent Team:
• Stage 1 - 45 mins with one of the team
• Stage 2 - Take home challenge
• Stage 3 - 60 mins technical interview with two team members
• Stage 3 - 45 min final with an executive and a member of the people team