Original listing text, shown exactly as published by the company.
The role
We're hiring an ML-focused Data Scientist to own retention and churn for CookUnity. In a weekly subscription business, retention is the engine of growth — the gap between a healthy customer and a churned one is often just a few weeks of activity. Your job is to predict who's at risk, understand why they leave, and power the interventions that keep customers ordering and win back the ones who go. This is a hands-on role embedded with Product, CRM, Marketing, and Engineering.
Responsibilities
- Churn & retention modeling: Build churn/survival models and lifecycle-state models that flag at-risk customers early and anticipate where each customer is heading.
- Intervention & reason understanding: Power the retention interventions that act on that risk — save flows, skip/pause deflection, lifecycle messaging — and classify why customers churn so the response fits the reason rather than being generic.
- Resurrection & win-back: Develop propensity models and personalized experiences that bring churned customers back.
- Personalization & Next-Best-Action: Decide the right action, offer, and message per customer across product surfaces, with an uplift layer measuring incremental impact.
- Offer & promo optimization: Determine who gets an incentive, when, and at what value — maximizing retained revenue without over-discounting.
- Production & MLOps: Own the full model lifecycle and ship models into the systems where they act; handle monitoring, retraining, and drift.
- Experimentation & incrementality: Design experiments and uplift measurement so we intervene on movable customers and not on those who'd retain anyway.
Qualifications
- 5-8+ years in data science, applied ML, or statistics, shipping production models.
- Retention/churn depth: churn prediction, survival / time-to-event modeling, and lifecycle-state models in a subscription or recurring-revenue context.
- Causal & experimentation rigor: uplift/incrementality measurement and A/B testing, with the judgment to separate true impact from selection effects.
- End-to-end ML & MLOps: building, validating, and deploying models in production (CI/CD, registries, containerization, orchestration, monitoring).
- Engineering & tooling: strong Python (pandas, scikit-learn, gradient boosting; deep learning a plus), SQL, code hygiene and reproducibility.
- Collaboration: excellent communication; able to embed with Product/CRM/Marketing and turn models into decisions.
- Education: BS in a quantitative field required; MS/PhD preferred.
- Ability to leverage generative AI to increase output quality and speed.
Preferred requirements
- Subscription marketplaces, food-tech, or consumer marketplaces with a retention mandate.
- Lifecycle-state / Hidden Markov models for churn.
- Causal and uplift libraries (e.g. EconML) or survival-modeling packages.
- Recommenders, embeddings, or personalization for retention and win-back.
Learn More About CookUnity
We believe great leadership starts with alignment on vision, values, and ways of working. To give you deeper insight into who we are and what we’re looking for, we invite you to explore: CookUnity's Leadership Principles – The values and behaviors that guide how we operate, collaborate, and scale.
We hope this provides valuable insight into our culture and product vision. If this excites you, we’d love to connect!
