until end of the year with option of extension
Zurich
Employment type
Contract
Industry
Banking
Area
Data Science
Location
Zurich
Remote from abroad?
No
Home office?
TBD
Contract duration
until end of the year with option of extension
01
Tasks and responsibilities
- This role will focus to redesign its current Client Risk Rating (CRR) model assigning clients to predefined tiers (e.g., Low, Medium, High).
- The goal is to refactor the model applying quantitative scoring architecture that enables granular differentiation between client risk levels.
- This scoring outcome will support both automated decision-making and human oversight across different Compliance and Financial Crime workflows.
- The candidate should leverage statistically sound yet interpretable methods, drawing from traditional risk scorecard design and lightweight classification models, ensuring high explainability.
- Prototyping and validation will occur in a secured sandbox environment with production-representative data.
- Refactor the existing categorical (Low/Medium/High) risk assignment process into a continuous or tiered numeric scoring system.
- Develop a points-based scoring engine inspired by best practices in credit risk and fraud detection, where each KYC attribute contributes a defined number of points according to its relative risk signal.
- Apply interpretable machine learning and statistical techniques suitable for regulated environments (for example Logistic regression with WOE or tree-based methods)
- Define and calibrate scoring bands ensuring clear thresholds and justifiable breakpoints.
- Build reproducible pipelines in Python for scoring calculation, backtesting, and sensitivity analysis, ensuring seamless handover to downstream teams.
- Full lifecycle ownership of analytical models including gathering business requirements, feature engineering, model development, validation, deployment, monitoring, documentation, and maintenance.
02
Must-have criteria
- Minimum 7 years professional experience in quantitative roles applying advanced data science and statistical modelling within top-tier banks/asset managers/fintech firms.
- Master's degree or PhD in Quantitative Finance, Mathematics, Physics, Engineering, or related field.
- Master-level Python command for data science (OOP, modular architecture, exception handling, performance profiling, test-driven development).
- Deep hands-on experience with pandas, NumPy, scipy, scikit-learn, statsmodels, PyTorch or TensorFlow.
- Proficiency in building reusable packages and internal libraries.
- Strong proficiency in SQL (Oracle).
- Full lifecycle ownership of analytical models.
- Exceptional analytical rigor.
- Outstanding communication skills.
03
Nice-to-have criteria
- Advanced academic training combined with real-world application in financial markets is highly valued.
04
Contract duration
- Start date: 01/09/2026
- End date: 31/12/2026
05
Language requirements
- English business fluent (C1/2)
- German basic knowledge (A1/2)
06
Application form