Main Responsibilities:
1. Algorithm Design & Prototyping: Design, develop, and validate predictive and analytical algorithms for CGM data. Develop robust code using advanced ML and statistical techniques to prove technical feasibility.
2. Feasibility & Ideation: Understand patient needs and creatively model potential algorithmic approaches using real-world sensor data.
3. Data Pipeline & Feature Engineering: Apply expertise in processing and managing heterogeneous time series data originating from medical devices. Execute rigorous data cleaning, imputation, transformation, and sophisticated feature engineering.
4. Technical Execution & Modeling: Build and optimize machine learning models (e.g., XGBoost, Neural Networks, etc.). Write high-quality, efficient, and reproducible Python code for data analysis, modeling, and experimentation.
5. Collaboration: Provide technical guidance within an Agile team framework to junior data science colleagues. Work effectively within a multidisciplinary, distributed team to translate project goals into actionable data science tasks.
6. Communication & Reporting: Synthesize complex technical results and present clear feasibility findings to diverse stakeholders.
Qualifications and Experience:
7. Relevant working/residency permit or Swiss/EU-Citizenship required
8. Minimum of 5+ years of hands-on experience as a Data Scientist or Machine Learning Engineer.
9. Demonstrated experience or robust academic background (Master or PhD is highly desirable) in Data Science, Machine Learning, Statistics, or a related quantitative field.
10. Strong Statistical Foundation: Solid grasp of statistical principles, experimental design, and model validation techniques.
11. Advanced Python Proficiency: Strong proficiency in Python and its core data science ecosystem: Pandas, NumPy, Scikit-learn, TensorFlow/PyTorch, and XGBoost/LightGBM.
12. Time Series Data: Practical experience with the processing, analysis, and modeling of time series data from physical sensors or monitoring devices.
Nice to Have:
13. Medical Domain Knowledge: Prior experience working with medical data, specifically in diabetes management (CGM/BGM), exercise physiology, or clinical nutrition data.
14. Regulated Environment: Familiarity with the requirements and processes for software development in a regulated medical device environment.
15. Big Data Tools: Experience with distributed computing frameworks like PySpark for handling very large datasets.