- Turn ambiguous clinical and operational questions into well-scoped ML tasks and build strong baselines.
- Build predictive models for member care, including disengagement risk, coach outreach, and health trends.
- Partner with product engineering to deliver model signals to members and care teams.
- Build model-specific production logic for retraining triggers, drift and anomaly detection, and continuous evaluation.
- Design offline and online evaluations, select metrics for imbalanced, delayed, and noisy healthcare outcomes, and analyze model failures.
- Build causal models validated against clinical experiments and partner with Product Analytics on CUPED/CUPAC variance-reduction techniques.
- Build ML- and GenAI-derived signals for personalized coaching and member experiences.
- Lead ML initiatives from problem framing through production, coordinating decisions and dependencies across teams.
- Share practices that strengthen model quality and delivery across the team.
Machine LearningNLP