Data Scientist
Summary
Build and optimize production fraud-detection models using Python, SQL, and Azure Data Lake to reduce false positives and improve accuracy.
Applied Data Scientist – Fraud Detection
Contract: 6 months
Engagement: Inside IR35
Location: Fully remote, with occasional travel to the London office.
The Role
We're looking for an experienced Applied Data Scientist to help improve the performance of our production fraud detection models.
Working within our Fraud Intelligence team, you'll analyse large-scale datasets, identify new behavioural signals and optimise machine learning models to improve fraud detection accuracy while reducing false positives.
This is a highly practical role suited to someone who enjoys working with complex data and delivering measurable improvements to live production models. You'll work closely with Machine Learning Engineers and Product teams to continually enhance the intelligence powering our fraud prevention platform.
Responsibilities
Analyse large-scale fraud and transaction datasets.
Engineer new features to improve fraud detection performance.
Optimise production machine learning models.
Improve model performance against key metrics including F1 Score, Precision and Recall.
Identify behavioural patterns associated with fraudulent activity.
Analyse behavioural biometric and device intelligence data.
Design and execute model evaluation experiments.
Build Python-based analytical workflows.
Write complex SQL against Azure Data Lake.
Present recommendations based on statistical analysis.
Work closely with Machine Learning Engineers to transition successful improvements into production.
Essential Skills
Commercial experience as a Data Scientist or Applied Data Scientist.
Strong Python and Pandas.
Advanced SQL.
Experience working with Azure Data Lake or similar cloud data platforms.
Experience working with very large datasets.
Strong feature engineering experience.
Experience improving production classification models.
Excellent understanding of F1 Score, Precision, Recall and model evaluation.
Experience working with highly imbalanced datasets.
Desirable
Fraud detection.
Financial crime.
Payments.
Banking.
Behavioural biometrics.
Device intelligence.
Transaction monitoring.
Risk scoring.