Medical Data Scientist – Evidence-Based AI for Longevity About the project: LAX is a longevity-focused health platform that uses AI to deliver evidence-based recommendations across three pillars: Lifestyle, Biomarkers, and Biohacking. In this initial phase, the goal is to build the scientific foundation of the AI system — a model that can evaluate, structure, and prioritize medical evidence (studies, guidelines, and meta-analyses) to generate transparent and reliable insights for preventive and longevity medicine. ⸻ Key Responsibilities • Develop and implement a scientific evidence-evaluation framework for the LAX AI • Integrate and process data from medical literature and clinical trials (e.g., PubMed, OpenAlex, ClinicalTrials.gov) • Apply natural language processing (NLP) and machine learning to extract and classify key findings from medical studies • Build a transparent evidence-weighting model (e.g., Meta-analyses > RCTs > Observational studies) • Collaborate with the LAX medical and research teams to define evidence criteria and data pipelines • Prepare data structures and algorithms that can later be scaled into deep learning models for personalization ⸻ Required Skills & Experience • Degree in Health Data Science, Medical Informatics, Biostatistics, or related field • Proven experience with medical or clinical research data • Proficiency in Python (Pandas, NumPy, Scikit-Learn, or PyTorch/TensorFlow) • Experience in data preprocessing, NLP, and evidence extraction • Understanding of clinical trial design and medical evidence hierarchies • Strong analytical and critical-thinking skills with scientific rigor ⸻ Nice to Have • Experience building knowledge graphs or evidence ontologies • Background in longevity, preventive medicine, or digital health • Familiarity with AI transparency and explainability in healthcare ⸻ Project Details • Remote position (part-time, approx. 2 h/day) • Duration: 2 months (with potential for extension) • Compensation: 62 CHF/hour (approx. 5000 CHF total) • Start date: As soon as possible