Master Thesis Student (m/f/d) – Explainable AI for Fraud Detection

The thesis will focus on Explainable Artificial Intelligence (XAI) for machine learning models in risk and fraud detection for online payments. A key focus will be on applying and evaluating Shapley value-based explanations (e.g., SHAP) to improve model transparency and validation, with an emphasis on making these methods robust and suitable for production use. The exact research question will be defined together with the student based on their interests and the team's priorities.

Qualifications

Your Profile

• Currently enrolled in a Master's program in Computer Science, Data Science, Mathematics, Statistics, Artificial Intelligence, Engineering, or a related STEM field.
• Strong programming skills in Python.
• Basic knowledge of machine learning and statistical modeling.
• Familiarity with Explainable AI concepts or a strong interest in model interpretability.
• Experience with Python machine learning libraries (e.g., scikit-learn, XGBoost, LightGBM) and data analysis.
• Familiarity with SQL; experience with Spark, Databricks, Docker, or cloud environments is a plus.
• Ability to work independently while collaborating effectively in an international team.
• Fluent in English.

Compensation: €960.00 / month

Master Thesis: Explainable AI for Machine Learning Models in Risk and Fraud Detection

Job Description

• Location: Berlin, Germany (Hybrid options available)
• Team: Data Science and Machine Learning (the team is English speaking)
• Employment Type: Full-time (dedicated to the thesis project)
• Starting Date: Flexible (Preferred start between September 2026 and March 2027)
• Duration: up to 6 months (standard German university thesis timeline)