Medicare Stars DI Data Scientist

At CVS Health, we’re building a world of health around every consumer and surrounding ourselves with dedicated colleagues who are passionate about transforming health care.

As the nation’s leading health solutions company, we reach millions of Americans through our local presence, digital channels and more than 300,000 purpose-driven colleagues – caring for people where, when and how they choose in a way that is uniquely more connected, more convenient and more compassionate. And we do it all with heart, each and every day.

Aetna, a CVS Health company, is the nation's premier health innovation company helping people on their path to better health. We are building a new health care model that is easier to use, less expensive, and puts the consumer at the center of their care.

Aetna's Medicare Stars Team is growing and expanding! This is exciting opportunity to join a high performing, collaborative team as a Data Scientist. This individual will focus on leveraging the Stars Data Ecosystem for Business Analytics for improved understanding of business patterns, predict future trends, and implement effective solutions to various challenges. This position will manage and be responsible for the successful delivery of scalable analytics solutions on Snowflake that combine structured and unstructured data, enable agentic document analytics at scale, and expose insights via natural language queries powered by LLMs (including OpenAI).

Own end-to-end implementation: ingestion, unified data modeling, vectorization/embeddings, retrieval pipelines, Snowflake compute (Snowpark), and integration with LLM-driven applications. This role will be utilizing their expertise in statistical analysis, machine learning, and data visualization to analyze complex data sets, then derive actionable insights that inform business strategies and decisions. By leveraging data, they help the Stars organization understand patterns, predict future trends, and implement effective solutions to various challenges.

• Design and implement unified data pipelines in Snowflake that combine structured tables, semi-structured data (JSON/Parquet), and large collections of documents (PDF, DOCX, text).

• Build agentic document analytics workflows: large-scale document ingestion, text extraction, cleaning, chunking, embeddings, vector stores, and efficient retrieval for analytic queries.

• Implement Natural Language Query (NLQ) interfaces that translate user text prompts into analytic queries or retrieval flows and return explainable results.

• Integrate Snowflake with LLMs (OpenAI or equivalent) for tasks such as summarization, question-answering, classification, and code-generation—using External Functions, Snowpark, and/or secure API patterns.

• Create performant retrieval-augmented generation (RAG) architectures that leverage Snowflake-stored embeddings and external or internal vector indexes as appropriate.

• Author Snowpark/Python/SQL transformations, Streams & Tasks, and job orchestration to enable near-real-time and batch analytical workloads.

• Implement data modeling and governance patterns within Snowflake: schemas, role-based access control, masking, lineage, and metadata to support analytics and compliance.

• Partner with product, ML/AI, BI, and engineering teams to translate business requirements into robust production-ready solutions.

• Build monitoring, observability, and cost controls for compute, storage, and API usage related to document analytics and LLM integration.

• Produce technical documentation, runbooks, and clear explanations of model/LLM behavior and limitations to non-technical stakeholders.

• Statistical Analysis:
  Proficiency in statistical methods and software, such as R, SAS, and Python, to analyze data.

• Data Visualization:
  Ability to create compelling visualizations using tools like Tableau, Power BI, and D3.js.

• Database Management:
  Knowledge of SQL and No
  
  SQL databases for efficient data storage and retrieval.

• Programming:
  Strong programming skills, particularly in languages such as Python, Java, and C++.

• Design and implement unified data pipelines in Snowflake that…