Sr Data Scientist

About the Role

• A health-tech company focused on longevity and preventive care is looking for a Health Data Scientist with deep statistical expertise to join their analytics team. This is not a standard data science role. You will work at the intersection of clinical evidence, actuarial thinking, and population health — building risk models that directly shape how the company identifies at-risk members, routes clinical interventions, and evaluates whether prevention programs are actually changing health trajectories.
• The questions you will work on are meaningful: Which members are most likely to develop a chronic condition in the next 12 months? What biomarkers from a longitudinal study best predict hospitalization? Are interventions moving the needle on health outcomes? If those questions genuinely excite you, this role was designed for you.

About You

• You are a statistician at heart who happens to work with health data. You understand that predicting member risk is ultimately about preventing a bad health outcome — not just estimating a cost. You are comfortable with uncertainty, rigorous about methodology, and able to translate complex quantitative findings into clear, actionable insights for clinical and business audiences. You come from a background in statistics, biostatistics, epidemiology, actuarial science, mathematics, quantitative biology, or public health with a strong quantitative component.
• What You'll Be Doing
• Build and maintain clinical risk models that generate predictive scores for hospitalization, chronic disease onset, and health deterioration — validated out-of-sample and monitored over time
• Conduct survival analyses of health events to help the clinical team determine when and how to intervene with specific member populations
• Design and execute statistical analyses that quantify the impact of prevention programs on health outcomes and claims frequency
• Develop risk segmentation frameworks grounded in biomarkers, diagnosis patterns, and longitudinal behavior — going beyond simple business rules
• Analyze claims patterns across risk cohorts, documenting frequency and severity differences by age band, condition, and risk profile
• Collaborate with clinicians, actuaries, and a Data Scientist to ensure models are grounded in clinical evidence and operationally actionable
• Deliver clear quantitative findings to clinical and business stakeholders through tables, charts, and written summaries
• Document methodologies so analyses can be reproduced, audited, and built upon over time
• What We're Looking For
• Someone who is genuinely fascinated by what data reveals about human health — not just someone who can run models
• A rigorous statistician who understands distributions, time-to-event outcomes, and how to quantify uncertainty in a clinical context
• A strong communicator who can make complex statistical findings accessible to clinical teams and non-technical leadership
• A collaborative professional comfortable working across disciplines — clinical, actuarial, and data engineering
• Someone with a track record of building models that hold up out-of-sample and can be explained and defended

Technical Requirements

• Must-Haves
• Statistical modeling in Python or R: GLMs, survival models, mixed effects models, and regularization techniques
• Survival analysis applied to health or clinical data: Kaplan-Meier, Cox proportional hazards, and accelerated failure time models
• Risk stratification or clinical segmentation using statistical or ML approaches
• Biostatistics or epidemiological methods: incidence rates, relative risk, hazard ratios, and propensity score matching
• Claims or health data analysis: frequency and severity patterns, cost drivers, and cohort differences
• SQL for data extraction and preparation from health or operational databases
• Ability to communicate quantitative findings clearly to clinical and business stakeholders
• Nice-to-Haves
• Experience with electronic health records (EHR), lab data, or clinical datasets
• Familiarity with actuarial concepts such as loss ratios, pricing basics, and experience studies
• Knowledge of causal inference methods: difference-in-differences, instrumental variables, or synthetic control for evaluating intervention effectiveness
• Proficiency with R packages such as survival , lme4 , or ggplot2 , or Python equivalents including lifelines , statsmodels , or scikit-survival
• Experience working with public health or insurance datasets in Mexico or Latin America