Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
Senior/Lead Data Scientist (Fintech): Building an analytical foundation to transform raw bank account data into reliable insights for fraud prevention and underwriting with an accent on feature engineering and ML model development. Focus on designing end-to-end production pipelines and developing fraud scorecards to estimate probability of default.
Location: Stockholm, Sweden
Salary: kr 775,398 – kr 1,108,832 SEK
Company
Klarna is a leading global fintech company specializing in smooth payment and shopping experiences.
What you will do
- Build the analytical foundation to turn raw bank account data into actionable insights for fraud prevention and underwriting.
- Perform feature engineering on incoming and outgoing cash flows to model affordability.
- Develop machine learning models to estimate probability of default and create fraud scorecards.
- Own the full model lifecycle, from raw data extraction through to production deployment.
- Collaborate with engineering teams to operationalize models at scale.
- Monitor model performance, impact, and acceptance rates via custom dashboards.
Requirements
- Experience building and deploying end-to-end machine learning models into production.
- Strong feature engineering skills, specifically with transactional or financial data.
- Proficiency in Python and SQL, with experience in AWS environments.
- Solid understanding of classification modeling techniques for risk or fraud use cases.
- Ability to work cross-functionally with Credit Modeling and Fraud teams.
- English: CV must be provided in English.
Nice to have
- Experience with open banking data or affordability modeling.
- Familiarity with Databricks for building model monitoring dashboards.
- Experience with observability tools such as Datadog.
- Professional background in underwriting, credit risk, or fraud detection.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →