ML Engineer

Описание

Sum Up provides simple and affordable financial tools that help small businesses manage payments, finance, and customer relationships. More than 4 million businesses in 37 markets rely on its products.

• Build and ship production ML systems end to end, including batch training pipelines, model versioning, monitoring, deployment, and rollback for transaction monitoring models
• Build, maintain, and improve ML models for transaction monitoring, focusing on detection quality, operational efficiency, and regulatory compliance
• Engineer features mapped to AML and Fraud typologies and suspicious behaviours, working with Risk investigators to translate domain knowledge into alerting logic and threshold calibration
• Run sensitivity tests on synthetic datasets and produce ML governance artefacts such as model cards and audit-ready documentation
• Own and evolve the AML Risk Score by analysing driver contributions, monitoring drift, running back-testing, and recommending improvements to features, logic, and thresholds
• Partner with AML and Fraud Operations, Product, and Engineering to translate stakeholder needs into actionable, scalable data science solutions
• Track and improve detection performance metrics, adapt solutions to regional compliance requirements, and contribute to system design documentation

• Write production Python code with CI/CD, automated testing, versioning, and monitoring experience
• Have experience deploying and operating ML models in production, including drift monitoring and rollback
• Train and product ionize ML models and choose appropriate evaluation KPIs and metrics
• Have hands-on experience with complex, multi-source data ecosystems, data quality, and data lineage
• Communicate clearly and confidently with cross-functional stakeholders, set expectations, and turn ambiguous compliance requirements into concrete technical plans
• Nice to have:
  Pyspark, AML, fraud detection or financial crime experience, unsupervised machine learning, Feature Stores, alerting threshold calibration, regulatory ML governance artefacts, AI systems and tooling

• Office-first setup in Berlin
• Virtual Stock Option programme
• Annual L&D budget of €2,000
• Corporate pension scheme matching up to 20% of contributions
• 28 Days of paid leave plus public holidays and special leave days
• Urban Sports Club subsidy, Kita placement assistance, and subsidised office lunches
• One-month sabbatical after 3 years of service
• Referral bonus