Data Scientist, Economic Crime Hub - NatWest Group

eFinancialCareers; Manchester; On-site; GBP 60,000 - 90,000; Full time

Job summary

eFinancialCareers is seeking a Data Scientist for our Economic Crime Hub in Manchester. You will design and implement data science solutions to prevent fraud, reduce customer harm and financial losses, and improve fraud decisioning.

You will collaborate with Fraud, Engineering and Data communities to promote data literacy, support governance and drive a data-driven culture across business stakeholders.

Qualifications

• Strong academic background in a STEM discipline such as Mathematics, Physics, Engineering or Computer Science.

Responsibilities

• Design and implement data science tools to harness data for fraud prevention and reduced financial losses.
• Collaborate with Fraud, Engineering and Data teams to identify opportunities and support strategic direction through better use of data.
• Translate fraud challenges into analytical questions and measurable outcomes, and deliver data-led solutions in an Agile environment.

Skills

Statistical modelling

Machine learning

Generative AI

Data visualization

Programming fundamentals

Education

STEM degree

Tools

Python

SQL

Cloud platforms

Job description

Join us as a Data Scientist, Economic Crime Hub

• You'll design and implement data science tools and methods which harness our data that use our data to prevent fraud and scams, reduce customer harm and financial losses, and improve the accuracy and efficiency of fraud decisioning
• We'll look to you to actively participate in the Fraud, Engineering and Data community to identify and deliver opportunities to support the bank's strategic direction through better use of data
• This is an opportunity to promote data literacy education with business stakeholders supporting them to foster a data driven culture and to make a real impact with your work

What you'll do

As a Data Scientist, you'll combine statistical analysis, machine learning, generative AI and software engineering to develop practical, responsible solutions to fraud and scam challenges. You'll work with fraud stakeholders and customer teams to understand their needs, form clear hypotheses and identify data-led solutions that improve fraud detection, reduce false positives, support timely intervention and deliver measurable fraud prevention and operational.

You’ll also be:

• Working with fraud stakeholders to translate fraud and scam challenges into clear analytical questions and measurable outcomes
• Applying a software engineering and product development practices to build reusable pipelines, test changes, and deploy scalable solutions in an Agile environment
• Selecting, building, training and testing machine learning models, fraud strategies and AI applications, balancing fraud detection and business value with customer impact, operational capacity, model risk and ethical considerations
• Monitoring internal and third-party fraud models for performance, data quality, drift and business effectiveness, recommending corrective action where needed
• Investigating emerging fraud patterns, unusual alerts and missed fraud events, turning findings into practical improvements and maintaining clear evidence for governance, audit and regulatory review

The skills you’ll need

You’ll need a strong academic background in a STEM discipline such as Mathematics, Physics, Engineering or Computer Science. You’ll have experience with statistical modelling and machine learning techniques applied to fraud or other complex risk problems involving rare events.

• The ability to use data to solve business problems from hypotheses through to resolution
• Experience using programming language and software engineering fundamentals
• Experience of Cloud applications and options
• Experience in synthesising, translating and visualising data and insights for key stakeholders
• Experience in model monitoring, model-risk governance and documenting analytical decisions for review and challenge is desirable.

• Knowledge of how Large Language Models and agentic AI can support fraud and scam analysis, and the controls required to manage the risks of using those applications, is also desirable