Data Scientist, Sprinter Health

• Deep experience building actuarial or health-economic models from administrative claims: total cost of care, PMPM, utilization, trend, and risk
• Command of the methods payers price on — MLR, risk adjustment, and multi-year projection — and the judgment to know their limits
• Strong SQL and Python or R, with the ability to build and own your models end to end
• Ability to hold your own with actuaries and medical-economics teams, and to explain the analysis to non-technical stakeholders
• Honesty about causal inference — what a given design can and cannot claim

Desirable

• We’re looking for a Data Scientist with an actuarial focus to quantify the long-term economic value of what Sprinter does
• When we tell a health plan that finding and treating disease in the home lowers their cost of care over time, someone has to turn that claim into numbers the plan’s actuaries will trust
• That is this role
• You will build the models yourself: total-cost-of-care and per-member-per-month (PMPM) projections from claims, medical-loss-ratio (MLR) impact, and a defensible account of how Sprinter’s interventions change cost and risk over a multi-year horizon
• You will package those outputs so a customer’s actuarial team can plug them straight into their pricing, reserving, and bid work
• This is a hands-on, applied role, not an advisory one — we would rather you build the table than write a memo about it
• The ideal candidate is a quantitative scientist who thinks like an actuary and is comfortable talking actuary-to-actuary with a payer’s medical-economics team
• Formal actuarial credentials are welcome but not required; we care that you can do the work and defend it
• This is a first-of-function role; you will define what good looks like
• Build total-cost-of-care, PMPM, and MLR models from claims data to quantify the long-term impact of Sprinter’s programs
• Project how interventions change cost, utilization, and risk over multi-year horizons, and quantify the uncertainty around those projections
• Produce model outputs and tables that a payer’s actuaries can plug directly into their pricing, reserving, and bid work
• Represent Sprinter in MLR and medical-economics conversations with health plans; go toe-to-toe with their actuaries
• Turn analysis into value narratives that quality, risk, and finance teams can act on
• Help the commercial team price and sell Sprinter’s impact on an actuarial basis
• Define the yardstick for whether an intervention actually changed cost and outcomes, not just whether it correlated with them
• Partner with Data Science on the causal and experimental design behind those measurements
• Bring an honest view of the line between value we can prove and value we can only assert

• We aim to complete the interview process between 2–3 weeks. It will usually consist of:
• Recruiter Screen (30 minutes)
• Hiring Manager Introduction (30 minutes)
• Hands-on-Keys Technical Assessment (1 hour)
• Onsite Interview: Systems Design / Technical Case Study + Research Presentation + Behavioral Interview + Lunch with the Team (4 hours)
• References