Staff Data Scientist

Midi Health•Palo Alto, CA

8h•$210,000 - $240,000•Hybrid

About The Position

We are looking for a highly strategic Senior or Staff Data Scientist to design, build, and own the end-to-end data framework that defines our business health: Unit Economics. In this role, you won't just build standalone models; you will connect the dots between customer acquisition, multi-product lifecycles, complex healthcare reimbursement cycles, and operational cost structures. Your work will serve as the financial and analytical source of truth, directly influencing how we allocate marketing spend, price our products, manage retention, and project long-term profitability. You will sit at the intersection of Data Science, Finance, Marketing, and Operations, acting as a critical strategic partner to executive leadership.

Requirements

Mastery of predictive modeling and Causal Inference techniques (e.g., uplift modeling, propensity score matching, synthetic controls, or diff-in-diff).
Proven experience architecture - building, deploying, and maintaining production-grade machine learning models. You write clean, modular, and well-tested code that integrates seamlessly into downstream workflows.
Deep expertise in model evaluation methodologies, backtesting, and validation. Because your models directly impact financial forecasts and pricing decisions, you have a rigorous approach to error analysis, cross-validation, and drift detection.
Proven track record building attribution models (algorithmic or heuristic) and handling survival analysis for churn and retention forecasting.
Advanced proficiency in Python for complex statistical analysis, alongside expert-level SQL for manipulating large data streams.
Experience structuring systemic business simulations or stochastic modeling.
Active adoption and mastery of Large Language Models (LLMs) and generative AI tools within your personal development workflow to accelerate coding, debugging, documentation, and prototyping.
You have a deep, near-obsessive understanding of the relationship between CAC, LTV, payback periods, gross margins, and contribution margins.
The ability to translate complex statistical outputs into clean, actionable frameworks for the CFO, CMO, and executive leaders. You know how to influence cross-functional roadmaps with data.
Ability to take vague, complex business questions and break them down into answerable, high-impact analytical components.
8+ years of experience delivering high-impact data science solutions.
Master’s or PhD in Economics, Econometrics, Applied Statistics, or a related quantitative discipline.
Demonstrated progression in scope and impact, with a history of acting as a strategic partner to finance and operations teams.
Candidates must be authorized to work in the U.S. without current or future sponsorship needs.

Nice To Haves

Ideally, your background includes time in Marketplaces, Healthcare operations, or D2C subscription businesses.

Responsibilities

Develop sophisticated lifetime value models that account for the volatility of healthcare reimbursements and the time value of money.
Build models to predict actual reimbursement rates across a complex mix of insurance allowables and self-pay tracks, closing the gap between theoretical revenue and cash-in-hand.
Establish the foundational frameworks that incorporate operational realities—such as state-by-state clinician licensing costs and wage ranges—ensuring our LTV calculations reflect true contribution margins.
Optimize "basket composition" and cross-sell dynamics between our physical supplement lines and clinical services to maximize total margin.
Build advanced attribution models (Markov chain, ML-based) to quantify the interplay between product lines—specifically tracking how supplement purchases drive clinical visit adoption and vice versa.
Design and analyze pricing experiments for supplement products to identify optimal margin-maximizing price points without degrading long-term subscriber retention.
Utilize your LTV and margin frameworks to influence the marginal LTV curves that marketing uses, helping them determine the exact point of diminishing returns on ad spend.
Build uplift models to identify which at-risk customers will respond positively to specific interventions (e.g., targeted offers, clinical outreach), preserving margin by avoiding unnecessary discounting on "sure things" or "lost causes."
Build stochastic (Monte Carlo) macro-simulations to help leadership and finance stress-test our business model. You will answer questions like: "If a major insurance payer shifts an allowable rate in a key state, how does that impact our payback period and portfolio margin?"