Sr Data Scientist

Join Us! HealthCare.com has become one of America’s fastest-growing insurtech companies, revolutionizing how consumers shop for health insurance. Leveraging advanced technology and data science, the company has developed customized proprietary products to better fit consumer requirements, enhance customer satisfaction, and take some of the guesswork and inefficiencies out of buying insurance. About the Role We’re hiring a Data Scientist (Production ML) to own and evolve our core production bidding system. This is not a research-only role and not a ticket-taking role. You’ll work directly with product, engineering, and business stakeholders to define problems, design solutions, and ship production-ready models that materially impact the business. We’re especially interested in people who are curious, pragmatic, and adaptable—the kind of data scientist who can move between modeling, data engineering, and software development as the problem requires. Prior experience matters, but how you think and learn matters more. This role sits on the data team alongside data engineers, data scientists, and analysts. What You’ll Do Own and operate machine learning models that run in production, including monitoring, debugging, and iterative improvement. Develop, train, and optimize models used in a real-time or near-real-time bidding and decisioning system. Work with stakeholders to clarify ambiguous problems, define success metrics, and translate business needs into technical solutions. Design and implement feature engineering pipelines, balancing model performance, latency, and maintainability. Write production-quality Python code (not just notebooks) and collaborate with engineering on deployment, CI/CD, and system design. Analyze model behavior using logs, metrics, and offline analysis to identify performance issues and opportunities. Contribute to data pipelines and infrastructure where needed (e.g., ETL, materialized tables, model inputs). Make thoughtful tradeoffs between something that is “theoretically optimal” and something that is reliable, fast, and shippable. What We’re Looking For Core Requirements Strong Python experience, including writing code that runs in production systems. Solid SQL skills and experience working with analytical databases (Snowflake or similar). Hands-on experience training and tuning tree-based models (e.g., LightGBM, XGBoost, CatBoost) on real, messy data. Experience deploying, maintaining, or owning ML models beyond experimentation (APIs, batch jobs, or streaming systems). Comfort working across dev, staging, and production environments. Ability to operate with limited guidance: you can ask good questions, propose solutions, and move work forward independently. Bonus Skills Experience at a smaller company or on a small team where you wore multiple hats. Depth in one or more of: Gradient boosted models and performance optimization Feature engineering for tabular data Data engineering / ETL design Model monitoring, evaluation, and debugging in production Experience improving model latency, reliability, or cost—not just accuracy. Prior technical leadership or informal mentoring experience. Familiarity with Airflow, Kubernetes, or cloud infrastructure. What This Role Is Not Not a research-only role. Not a role where requirements are always fully specified up front. Not a notebook-only environment. Not a position where “training a model once” is considered done. Why You Might Like This Role You’ll own systems that directly affect revenue and business outcomes. You’ll have real autonomy over technical decisions. You’ll work on problems where good judgment matters more than shiny tools. You’ll see the full lifecycle of models—from idea to production to iteration.