Data Scientist

Role overview

This is a foundational Data Scientist position on a small data team, focused on putting machine learning into production rather than notebooks. The role partners closely with Product, Engineering, and Lifecycle Marketing to ship models as live product features, starting with a personalization model that decides what offer to show each user at the paywall and across the lifecycle. The hire will also help establish the team's first real-time, low-latency serving patterns on GCP.

Responsibilities

- Build, validate, and deploy ML models (propensity, pricing and discount optimization, personalization, churn and LTV) that run inside the product, end-to-end from problem framing to monitoring and retraining. - Design and ship the personalized discounting model, including experiment design and a measurement framework that holds up to finance and leadership scrutiny. - Stand up the first low-latency model serving pattern on GCP (Vertex AI endpoints, Cloud Run, or equivalent) and define how features flow between batch warehouse tables and real-time streams. - Implement monitoring for drift, staleness, and prediction quality so live models do not silently degrade. - Translate product problems into modeling problems, then into clear API contracts engineering can own long-term. - Design causal and uplift models and run controlled experiments that prove incremental lift on revenue and retention.

Requirements

- 4-7 years in a data scientist or ML engineer role, with at least one model personally taken from prototype to live production serving real traffic. - Strong Python for ML, including scikit-learn and gradient boosting libraries such as XGBoost or LightGBM. - Hands-on experience with a cloud ML platform, ideally GCP (Vertex AI, BigQuery ML, Cloud Run or Functions), or the ability to translate equivalent AWS or Azure experience quickly. - Solid SQL skills and comfort working in a dbt and BigQuery warehouse. - Software engineering fundamentals: git, code review, testing, and CI/CD, with code reviewed by engineering peers. - Practical causal inference or uplift modeling experience applied to pricing, discounting, or similar problems. - Quantitative background in computer science, statistics, engineering, or equivalent hands-on work; fluent English.

Nice to have

- Experience with streaming or event pipelines such as Pub/Sub, Dataflow, or Kafka. - Direct experience with pricing, discounting, or personalization use cases. - Familiarity with feature stores or DIY versioned feature pipelines. - Multi-armed bandits or reinforcement learning applied to pricing or personalization. - Background in B2C, subscription, or marketplace products, or early-stage startup environments where the first ML pattern is being defined from scratch.