Applied ML Engineer / Data Scientist

About the role

grape is an employee insurance company looking for an applied ML engineer / data scientist to build predictive models from the ground up on real, regulated insurance data. The role owns problems end to end, from effective-dated records to production, in areas such as pricing and risk, claims, triage, return-to-work and fraud detection. It is not an LLM role: LLM and agent workflows sit with the engineering teams.

Key responsibilities

• Build and validate models, owning the full lifecycle: feature engineering, training, evaluation, explainability, deployment and drift monitoring.
• Backtest on historical data and run models in shadow mode alongside human decision-makers, reporting disagreements.
• Produce audit-ready validation and governance documentation for actuarial, compliance and carrier partners.
• Deploy models on the company's cloud platform within strict security and data-residency requirements for health data.
• Set up experiment tracking, model versioning and production monitoring, partnering with engineering.
• Identify and prioritize high-value use cases, turn ambiguous questions into measurable modelling problems, and embed privacy, fairness and explainability in line with nDSG/GDPR.
• Collaborate with underwriting, claims, actuarial and compliance teams on model behaviour and limits.

Required profile

• Strong applied ML and statistics, including supervised and unsupervised learning, anomaly detection, time-series and survival/duration modelling.
• Experience building models on real-world, relational, effective-dated data, including imbalanced outcomes.
• Fluent English; German is a big plus.
• Big plus: insurance, actuarial or risk modelling; Swiss insurance data (KTG, UVG); regulated special-category data under nDSG/GDPR; Google Cloud.

Required skills

• Fluent Python (pandas, scikit-learn, a modern ML stack) and strong SQL.
• Solid grasp of validation, calibration and explainability (e.g. SHAP).
• Production MLOps: experiment tracking, versioning and drift monitoring.

What we offer

• Hybrid work with a team day in the Zurich office, plus a CHF 200 home-office budget and a monthly internet allowance.
• CHF 1,500 annual learning budget and mental health sessions.
• Flexible working hours, 27 paid vacation days per year, and a CHF 150 SBB half-fare card.
• Wellness budget, parental leave, and digital nomad option after 3 years.