Staff Machine Learning Scientist

What You'll Accomplish

• Send-time and channel optimization: Design and ship the next system for deciding what nudge to send a member, when, and through which channel, beyond our current contextual-bandit approach.
• Propensity modeling: Build and deploy models that decide whether nudging a given member is worth it, balancing engagement against fatigue and unsubscribes.
• Experimentation rigor: Set the bar for how the team runs experiments: multi-arm tests, sequential testing, CUPED, and guarding against peeking, so our nudge decisions are causally sound.
• Production ownership: Own at least one model in production end-to-end.
• Leadership: Mentor the team's ML scientists, guide technical direction, and partner across product, engineering, data science, and the growth and marketing teams.

Required Qualifications

• Bachelor's degree or higher in Computer Science, Statistics, Operations Research, Machine Learning, or a related quantitative field
• 7+ years building and deploying ML systems in production at consumer scale
• At least one recommendation, ranking, or sequential-decisioning system shipped end-to-end (modeling, evaluation, deployment, monitoring, iteration)
• Fluency in experimentation and A/B testing: multi-arm tests, sequential testing, CUPED, and the common failure modes of online experiments
• Proficiency in Python and SQL; able to read a colleague's PR and improve it
• Deep understanding of machine learning and applied statistics

Preferred Qualifications

• Contextual bandits or reinforcement learning operated in production
• Multi-objective optimization (engagement vs. adherence vs. retention vs. cost)
• Causal inference beyond A/B testing: difference-in-differences, synthetic controls, instrumental variables
• Cold-start and low-data-regime modeling (healthcare gets thin on per-member data fast)
• Experience hiring and growing a small ML team
• Healthcare, fintech, or other regulated-data experience; familiarity with HIPAA and BAA constraints
• Familiarity with our adjacent stack: Statsig, Databricks, feature stores, Airflow/dbt
• Familiarity with TypeScript