Data Scientist

About the Role

As a Data Scientist at Ophelia, you'll build the statistical and machine-learning models that turn our data into forward-looking decisions, including building forecasting models for business operations, as well as marketing models to drive conversion. You'll take open operational and growth questions and turn them into rigorous, well-scoped models that ship and get used.

You will collaborate across teams to identify data-driven opportunities that propel the business forward, delivering high-quality work using an agile methodology. Your contributions will ultimately influence key company domains, including company-level KPIs, clinical, commercial, and business operations, and more. Data is deeply embedded in our corporate culture, with widespread adoption of data products and significant potential for further evolution.

In this fully remote position, you will report to our Senior Director of Data and serve as a peer to our analysts and analytics engineers. As our first dedicated data scientist, you'll have the ability to shape how modeling, experimentation, and machine learning are practiced here - spanning clinical-capacity forecasting, no-show and retention prediction, outcome and risk models, and NLP on clinical documentation. You'll partner closely with Analytics Engineering to move models from notebook to production, and with the broader Technology Team to cultivate a culture of data-driven decision-making that ultimately enhances patient care.

Ophelia's Technology team actively encourages and invests in AI-augmented ways of working: using tools like Claude and Gemini to accelerate coding, analysis, and experimentation, and staying current as the AI landscape evolves. Consistent with Ophelia's AI Position Statement, we treat AI as a force multiplier for our team, not a replacement for human judgment - it's there to help you work at the top of your license, move faster from question to insight, and spend more time on the modeling and reasoning that actually requires a person. You'll be expected to use AI fluently in your own workflow, and to help shape how the larger technology team uses it responsibly and effectively as it evolves.

Together, we are dedicated to helping individuals with opioid dependence. While experience in this specific treatment area isn't required, ideally you will possess knowledge of the healthcare landscape, including an understanding of health outcomes, benchmarks, systems, and regulatory compliance (HIPAA).

Responsibilities

• Predictive & statistical modeling: Build, validate, and ship models - demand and clinical-capacity forecasting, no-show and retention risk, outcome and risk prediction - that directly inform clinical and business strategy.
• Experimentation & causal inference: Design and analyze experiments and quasi-experiments so the organization can measure true impact, not just correlation.
• Productionalizing models: Collaborate with Analytics Engineering and Technology teams to deploy models into robust pipelines (BigQuery, dbt, Dagster) with continuous monitoring, retraining, and drift detection - ensuring models operate as living products.
• From ambiguity to rigor: Translate open clinical and operational questions into well-scoped analyses that are mathematically sound and clearly caveated.
• Cross-functional partnership: Serve as the modeling partner for Clinical, Commercial, and Business Operations - surfacing high-leverage opportunities and communicating results in plain language.
• Raising the bar: Act as the subject-matter expert for statistical and ML methods, championing rigorous, evidence-based decision-making across the team.

Skills & Competencies

• Statistical & ML depth: 3-5+ years of applied data science with a strong foundation in statistics, probability, experimentation, and machine learning. You know when a simple model beats a complex one and how to distinguish signal from noise.
• Python for data science: Advanced proficiency in the Python ML/analysis stack (pandas, scikit-learn, statsmodels, gradient-boosting frameworks such as LightGBM/XGBoost, and forecasting methods). Familiarity with R is a plus.
• Operations research: Familiarity with resource-allocation, scheduling, or optimization problems (e.g., linear/integer programming, queuing theory, simulation) is a strong plus - relevant to how we plan and allocate clinical capacity.
• AI-augmented workflow: Comfortable and enthusiastic about using modern AI tools (e.g., Claude, Cursor, Copilot) to accelerate coding, analysis, and research, while applying the judgment to know when human review and validation matter most.
• Communication & influence: You turn complex modeling into clear narratives and "so-what" recommendations that a non-technical audience can act on.
• Analytical versatility: You thrive in environments where data is semi-structured and projects span a wide variety of business domains, and you navigate ambiguity comfortably.
• Productionalization: Interest in taking models from exploration into production pipelines with monitoring and retraining; orchestration experience (Dagster, Airflow) are a plus.
• Mission & industry alignment: A passion for making evidence-based addiction treatment accessible. Experience in a regulated industry (healthcare, fintech, etc.) is a plus.

#LI-Remote