Lead Data Scientist — End-to-End Fraud ML

Deel; Greater London; Híbrido; GBP 81.000 - 116.000; Tempo integral

Resumo da vaga

Klarna in London is seeking a Senior/Lead Data Scientist to own end-to-end ML pipelines for fraud detection, from raw data to production models. You will rebuild detection systems at scale across hundreds of millions of transactions, converting rules-based approaches into model-driven architectures.

You will work with engineers and stakeholders to ship real-time inferences, deploying novel techniques such as graph networks and anomaly detection into production where they create real impact.

Qualificações

• End-to-end ML ownership across data engineering, feature development, model design, training and production deployment.
• Strong instinct for when a model is ready for production and when it is not.
• Proven track record of building ML models and pipelines from scratch, not just integrating tools.
• Experience building real-time or near-real-time inference systems.
• Comfortable with large-scale datasets including hundreds of millions of transactions and high-dimensional features.
• Excellent communication with senior stakeholders.

Responsabilidades

• Build fraud-detection ML systems from data to production at scale.
• Translate business problems into precise technical solutions.
• Collaborate with engineers, analysts, and commercial stakeholders.
• Incorporate novel approaches such as graph networks and anomaly detection into production.

Conhecimentos

Python

SQL

scikit-learn

LightGBM

Formação

Degree in computer science / physics / mathematics / engineering / related quantitative field

Ferramentas

Docker

Jenkins

AWS

SageMaker

Lambda

S3

Athena

Descrição da vaga

Klarna in London is seeking a Senior/Lead Data Scientist to own end-to-end ML pipelines for fraud detection, from raw data to production models. You will rebuild detection systems at scale across hundreds of millions of transactions, converting rules-based approaches into model-driven architectures.

You will work with engineers and stakeholders to ship real-time inferences, deploying novel techniques such as graph networks and anomaly detection into production where they create real impact.