Data Scientist

Hospital for Special Surgery•New York, NY

About The Position

Hospital for Special Surgery (HSS) is hiring a Data Scientist to work with domain experts across clinical, operational, research, finance, and executive teams. The role turns ambiguous needs into well-defined problems, with explicit assumptions, actionable decisions or workflows, and credible measures of impact and value. The Data Scientist develops advanced analytics and data-science-driven features for applications and products that users can rely on, rather than delivering stand-alone notebooks. The work can span design, prototyping, development, testing, deployment, integration, monitoring, and iteration. Methods may include predictive modeling and forecasting, experimentation and A/B testing, causal or program-effectiveness analysis, segmentation, anomaly detection, risk stratification, and measurement design. The position also involves productionizing analytical work, integrating models and logic into applications and Databricks workflows, evaluating and monitoring performance, responding to drift or failure, and improving products after launch. The Data Scientist contributes to reusable capabilities such as model-development and evaluation pipelines, model cards, monitoring, documentation, and reproducibility. Work is collaborative and includes synchronous working sessions with domain experts, clear communication of quantitative concepts, and judicious use of AI coding tools.

Requirements

Bachelor’s degree in a quantitative field or equivalent demonstrated technical ability.
Strong foundations in data science, statistics, and modeling, including the ability to explain and defend methodological choices, identify leakage and confounding, quantify uncertainty, and select problem-appropriate performance metrics.
Strong programming ability and experience, such as Python and SQL.
Comfort working with git, building within applications, and working in a cloud or lakehouse environment.
Experience with projects involving large datasets and the design, prototyping, development, testing, deployment, and monitoring of statistical and machine-learning models.
Flexibility, communication, collaboration, comfort with ambiguity, and ability to work with domain experts who are not data experts and translate complex concepts for them.
Drive, ownership, and a bias toward action, including the ability to form a point of view about where work should go.
For Data Scientist I: generally 0–2 years of professional data science or advanced analytics experience; strong graduate, internship, fellowship, research, or substantial applied-project work may count. Candidates should have exposure to end-to-end production work or be ready to learn it quickly.
For Data Scientist II: generally 3–5+ years of professional data science or advanced analytics experience, with a record of independently leading work from problem discovery through deployment, application integration, monitoring, and iteration. Substantial end-to-end production experience is required.
For Data Scientist II: ability to operate autonomously, lead through influence, manage competing priorities, mentor others, and connect work across domains.

Nice To Haves

Master’s degree or above; preferred but not required for Data Scientist II roles.
Prior Databricks experience.
Experience building data-science-driven product features or decision-support tools from problem ideation through deployment, monitoring, and iteration.
Experience in forecasting, experimentation, causal inference, anomaly detection, risk stratification, optimization, or measurement design.
Experience with healthcare data in a U.S.-based healthcare delivery or health-tech setting, especially musculoskeletal health and orthopedics.
Experience in operations, finance, economics, research, clinical science, consumer or digital products, or other complex enterprise settings.
Strong experience with modern AI agentic coding tools and techniques, including judgment about which tasks to delegate and how to verify results.
Curiosity about current AI models, their strengths and weaknesses, and effective ways to use them.

Responsibilities

Work with domain experts to identify opportunities and turn vague questions into well-posed problems with explicit assumptions, actionable decisions or workflows, and credible measures of impact and value.
Build data-science-driven features for applications and products that clinicians, operators, researchers, finance teams, and executives can use.
Contribute across product design, prototyping, development, testing, deployment, integration, monitoring, and iteration.
Choose and apply appropriate analytical methods, including predictive modeling, forecasting, experimentation, causal analysis, segmentation, anomaly detection, risk stratification, and measurement design.
Challenge weak analytical designs, quantify uncertainty, and use simpler methods when they better fit the decision.
Write maintainable code and integrate models and analytical logic into applications and Databricks workflows.
Create appropriate model evaluation and monitoring, respond to drift or failures, and improve products after launch.
Develop reusable approaches for model development, feature and evaluation pipelines, model cards, monitoring, documentation, and reproducibility.
Identify opportunities to reuse methods, datasets, features, or workflows across domains.
Collaborate with domain experts in daily synchronous working sessions, ask questions, adapt to evidence, and explain complex quantitative ideas clearly to varied audiences.
Use agentic coding tools thoughtfully, deciding what to delegate and checking for errors and implicit assumptions.