Senior/Lead Data Scientist-Fraud
Pubblicato il 16-08-2026 - Klarna in Milano
Job ID: 5d4ec96e-7eb2-4248-8f04-849cd040dced | Location: Milan | Salary: 63236 | 90429 EUR | Type: Tempo pieno | Posted:
Di seguito troverà tutto ciò che deve sapere su questa opportunità e sui requisiti richiesti ai candidati.
We are rebuilding our fraud detection systems from the ground up, and we are looking for data scientists who want the full mandate to do it right, from raw data to production model, at a scale that affects hundreds of millions of transactions.
Klarna is a big company that still moves like a startup: fast decisions, real ownership, and models that ship. We own the full stack, and expect you to build it with us to enable your use cases. If you have the curiosity to find the right problem, the grit to build the right solution, and the ambition to see it matter, this is the opportunity.
Fraud is one of the most technically demanding problem spaces at Klarna.
Key responsibilities
- Build best-in-class machine learning systems from the ground up, owning the complete pipeline from raw data through feature engineering, model design, training, and real-time production deployment.
- Convert Klarna's existing rules-based fraud systems into sophisticated, model-driven architectures that operate at scale across hundreds of millions of transactions.
- Build infrastructure from scratch, not maintain or extend existing frameworks.
- Translate ambiguous business problems into precise technical solutions and bring novel approaches such as graph networks, anomaly detection, and behavioural signals into production where they create real impact.
- Work with engineers, analysts, and commercial stakeholders.
Who you are
- End-to-end ML ownership across the full stack: data engineering, feature development, model design, training, low-latency production deployment, monitoring, and retraining.
- 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 integrating or extending someone else's product or tooling.
- Experience building real-time or near-real-time inference systems; batch pipelines alone are insufficient.
- Comfortable with large-scale datasets including hundreds of millions of transactions and high-dimensional feature spaces.
- Strong Python and SQL skills with hands-on experience in scikit-learn, LightGBM, Docker, Jenkins, and modern Python packaging.
- Self-motivated, fast-moving, and creative.
- Bring novel solutions where others reach for off-the-shelf tooling.
- Communicates precisely across technical and non-technical audiences including senior stakeholders.
- Degree in computer science, physics, applied mathematics, astrophysics, automatic control, mathematics, software engineering, electrical engineering, or a related quantitative field.
- Awesome to have: Experience building end-to-end ML systems in early-stage startups or small greenfield teams.
- Hands-on production experience with graph neural networks, anomaly detection, or behavioural biometrics, beyond prototyping or fine-tuning.
- Familiarity with AWS (SageMaker, Lambda, S3, Athena) and CI/CD practices. xysqume
- Experience mentoring or technically guiding other data scientists.
#J-18808-Ljbffr