Our team works AI-first. You will work with large, imperfect consumer datasets, use AI tools daily as a development partner, and turn model output into recommendations the business can act on. You will partner closely with data engineering, product and marketing.
What You’ll Do
Predictive Modeling
- Own the modeling roadmap for your area, from problem framing through production and ongoing monitoring
- Build cohort and LTV models that inform user acquisition budgets and payback decisions
- Build user segmentation and propensity models that support personalization and targeting
- Improve model accuracy over time through validation, backtesting and new data sources
- Design experiments and causal analyses that measure the true impact of product and marketing changes
- Identify early behavioral signals that predict conversion, churn and long-term value
AI & Data Products
- Apply LLMs and machine learning to classify unstructured text and behavioral signals into interest categories (such as the IAB taxonomy) and product and brand intent, powering personalization and audience targeting
- Partner with data engineering to productionize models and keep their inputs reliable
- Build internal AI-powered analytics tools and data products
Communication
- Explain model results, assumptions and confidence levels clearly to stakeholders and leadership
- Set modeling standards, documentation and best practices for the team
- Mentor other analysts and data scientists and share best practices across the team
Who You Are
- 5+ years of experience in data science, statistical modeling or machine learning, ideally with consumer app or web data
- Strong Python (pandas, scikit-learn or similar) and advanced SQL
- Experience deploying models to production and monitoring them over time
- Solid grounding in statistics, forecasting and experimentation
- Hands-on experience with LLM APIs, prompt design or building AI-powered data tools
- Daily use of AI tools as a development partner, such as Claude, Claude Code or Claude Cowork
- Sound judgment with incomplete or noisy data; you structure the problem before reaching for a tool
- Strong communication skills, including turning complex analysis into clear business recommendations
- Fluent English and at least 4 hours of overlap with US Eastern business hours (remote roles)
Nice to Have
- Experience with ad-monetized or subscription consumer apps
- Experience with recommendation systems, personalization or audience segmentation for ad targeting
- Experience with ML platforms on AWS or Google Cloud, such as SageMaker or Vertex AI
- Familiarity with MLOps and orchestration tools such as MLflow, Airflow or dbt
- Familiarity with Bayesian methods, uplift modeling or multi-armed bandits