Senior Data Scientist

Solaris is Europe’s leading embedded finance platform, pioneering Banking-as-a-Service to bridge the gap between financial technology and licensed banking. Founded in 2016 and headquartered in Berlin, Solaris enables fintechs, digital ecosystems, and multinationals to embed accounts, cards, lending, and payments directly into their own products, creating seamless financial experiences for their users. As a fully licensed German bank, Solaris combines the highest standards of IT security and regulatory resilience with a modular, highly scalable infrastructure.

Today, Solaris is driving the next evolution of financial services by transitioning into an AI-native banking platform, ensuring European businesses can deploy secure, adaptable financial products at scale.

• Development, operationalisation and maintenance of Machine Learning models in close collaboration with the business stakeholders for common risk and financial protection with different latency: Fraud Protection, Compliance & AML
• Training data preparation in close collaboration with the analytics engineers including analysis of vast amounts of transactional logs, data labelling, applying chronological splitting and sampling techniques to handle class imbalances
• Feature engineering operations including common features, cross features, positional features and building a centralised feature store
• Model selection, experimentation  and training of baseline and gradient boosted models, evaluating performance and trade offs
• Model deployment and prediction servicing from batch to online in close collaboration with the data infrastructure team
• Continual learning by setting up automated pipelines that monitor population drift and continuously re-fit models on fresh data when performance drops below predefined operational baselines.
• Knowledge sharing and mentoring across the team.

Depending on your level of experience, your responsibilities and scope of role will range. We don’t care much about fancy titles, but rather about real personal and professional development, as laid out in our learning framework. Let’s figure together out how you can contribute to our team.

• Degree in Computer Science, Applied Mathematics, Statistics, Quantitive Finance and targeted Financial Engineering courses.
• Minimum 6 years experience in a role of data scientist in a fast pace environment and regulated industry.
• Proficiency in data science libraries (pandas, polars, numpy, scikit-learn) and gradient boosting frameworks (XGBoost).
• Advanced SQL skills (window functions, query optimisation) and hands on experience in analytical platforms (ideally Snowflake by utilising snowpark). 
• Experience working with centralized feature platforms (e.g., Snowflake Feature Store, Feast, Tecton) to prevent train-serve skew.
• Good knowledge of data transformation and data orchestration tools, ideally dbt and airflow.
• Solid understanding of software engineering principles, including version control (Git), CI/CD, and automated testing.
• Payment, Fraud and Risk Domain Expertise:
  • Understand transactions movement, payment payload and authentication protocols.
  • Recognize differences in typologies, spotting anomalies and understanding chargeback and dispute cycles
  • Velocity Features, Device Dynamics and Entity Profiling
  • Financial and Regulatory Guardrails
• Excellent English language skills, and preferable German language skills.
• Very strong communication skills, to both technical and non technical members and ability to explain complex statistical outputs to non technical officers. 
• Ability to grasp new business concepts and translate them into technical requirements.
• Crisis communication under pressure in periods of unplanned situations.
• Adaptability and continuous learning.
• Adversarial & Skeptical mindset.
• Ability to mentor and inspire team members.
• Comfortable working with AI tools, thinking critically about AI outputs, and contributing to a culture of responsible AI use. We expect you to demonstrate comfort with AI-assisted workflows and a willingness to continuously develop their AI capabilities as the technology evolves.