About the role

Senior Data Scientist at Hungryroot, owning machine learning to personalize customer deliveries. Collaborates within a remote-first team to optimize data models and improve customer satisfaction.

Responsibilities

Separate durable preference from noise. Design robust feature representations from high-cardinality, implicit behavioral data (swaps, skips, saves) to capture true user intent and predict future engagement.
Model temporal dynamics and changing tastes. Architect sequential and recency-aware systems that adapt to shifting user preferences, ensuring recommendations reflect current intent rather than stale history.
Solve the cold-start problem. Leverage cohort signals, clustering, and content embeddings to generalize learnings across users, ensuring that even a new customer’s first box feels deeply personalized.
Bridge ML and constrained optimization. Integrate model scores (e.g., predicted conversion) with operations-research engines to perform business-aware re-ranking, balancing personalization with hard constraints like diet, budget, and inventory.
Advance the modeling. Evolve our systems using the architectures that drive modern, high-scale personalization, such as multi-stage retrieval and ranking, learning-to-rank (LTR), matrix factorization, and gradient-boosted trees. You will also evaluate and integrate more sophisticated techniques (like contextual bandits or sequence modeling) as our data complexity grows.
Drive rigorous experimentation. Define robust offline evaluation metrics (e.g., NDCG, MAP) and design online A/B tests to measure true causal impact on customer retention and satisfaction

5+ years of hands-on experience in data science, applied machine learning, or a related quantitative role.
Champion ML system best practices. You treat the ML lifecycle as a rigorous discipline, moving systematically from problem definition and feature engineering to robust offline evaluation, online experimentation, and CI/CD for ML.
Deep expertise in personalization, search ranking, or recommender systems, with hands-on experience building multi-stage architectures (candidate generation, scoring, and re-ranking).
Strong grounding in statistics, causal inference, and experimentation, with the ability to define proxy metrics and design tests that measure long-term business impact.
Production-level engineering skills in Python and SQL, with hands-on experience scaling models using big data frameworks and an understanding of system latency trade-offs.
A commercial mindset to translate complex business constraints into scalable ML architectures.
Clear communication and a collaborative, remote-friendly working style, including mentoring others.

Remote-first: work from home, work from our NYC office, work from anywhere in the U.S. - you decide!
Equity
Unlimited vacation policy
Universal paid parental leave
Monthly Hungryroot credit for delicious, healthy groceries
Comprehensive health, vision, dental, and life insurance
401k with Company Match
A work from home stipend to support your initial home-office setup