Finde deine nächste berufliche Chance

Virtusan is a Swiss based health coaching platform for sleep, stress and recovery that works with any wearable, including our own smart ring. We're hiring our first AI Tech Lead to build the intelligence behind our coach: turning biometric and behavioural signals into personalised programmes that help people sleep better and stress less. You’ll be the technical authority for AI across engineering; you make the calls on architecture and approach, set the standards others build to and stay hand-on in the code. What you'll be doing 1) Signal interpretation: Turn wearable and in-app data into recovery, stress and sleep insights, using personal baselines and trend detection. 2) Multi-device data: Normalise metrics across vendors, weight sources by reliability, and handle missing data and device switches. 3) Recommendation engine: Personalise protocol sequences by dose and timing with classical ML (e.g. contextual bandits, sequential recommenders or RL). 4) Feedback loops: Learn from real user behaviour and biometric responses which protocols work, and for whom. 5) Evaluation: Offline metrics, online experiments and production monitoring for every model and LLM feature. 6) AI platform: Own the roadmap, training and serving pipelines, data quality and architecture on our Python and Azure stack. 7) Technical leadership: Make and own the key architecture and modelling decisions, and write them down (RFCs, ADRs etc.) so the team can build them. 8) Privacy: Make secure handling of sensitive health data part of the design from the start. 9) Collaboration: Help product, design and marketing shape AI features and understand what the models can and can't do. Hands on: This is a builder role. You’ll be prototyping, evaluating and taking code through to production. Requirements 6+ years in ML/AI engineering incl. 2+ years in a lead role, with at least one product taken from zero to production Python, FastAPI and production MLOps on Azure or another major cloud Time series, recommender systems, causal/uplift methods and strong statistics A/B testing, model evaluation and production monitoring LLMs and LangGraph (or similar) on Azure OpenAI, and knowing when not to use them Experience as a tech lead or staff engineer, making technical decisions that a team of engineers followed. Privacy-aware handling of health data Based in Europe and fluent in English Nice to have: bandits, RL or off-policy evaluation; health or wearable data; literature mining; AI in engineering workflows (coding agents, AI-assisted testing, CI). Application process Instead of a cover letter, please send us the answers to 2 of the following 3 questions together with your CV. 1) Tell us about an ML product you took from zero to production. What did you own, and what changed because of it? 2) You have a recommendation engine and want to know whether a protocol actually improves users' sleep. How would you measure that? 3) Tell us about a technical decision you made for a team that others disagreed with. How did you get alignment, and how did it turn out?