Software Development Engineer

We’re building a world of health around every individual — shaping a more connected, convenient and compassionate health experience. At CVS Health®, you’ll be surrounded by passionate colleagues who care deeply, innovate with purpose, hold ourselves accountable and prioritize safety and quality in everything we do. Join us and be part of something bigger – helping to simplify health care one person, one family and one community at a time.

Position Summary
AI LaunchPad is CVS Health's centralized AI platform for the HCD Business Unit. We provide LLM access, observability tooling, and the infrastructure needed to build and deploy production-grade AI agents across the enterprise. Our engineers work at the intersection of cutting-edge generative AI research and real-world healthcare impact — shipping agentic systems, RAG pipelines, and multimodal workflows that touch millions of patients and providers.

Role
We are looking for a Software Engineer with a strong Generative AI and LLM engineering background to join the AI LaunchPad team. You will design, build, and operate production AI systems — from agentic orchestration pipelines and RAG architectures to LLM evaluation frameworks and MLOps infrastructure. You will have a direct hand in reducing the POC-to-production cycle time and accelerating AI adoption across HCD.

• Design and deploy enterprise-scale Generative AI platforms including agentic RAG pipelines, multimodal workflows, and LLM-powered automation in HIPAA-compliant cloud environments.

• Build and maintain agentic orchestration systems using frameworks such as LangGraph, LangChain, CrewAI, or Google Agent Development Kit (ADK).

• Develop and operationalize RAG systems with advanced retrieval techniques (hybrid search, reranking, query rewriting) and robust evaluation pipelines.

• Establish LLM observability and CI/CD guardrails — integrating tools like LangFuse and RAGAS — to enable prompt regression detection and production stability.

• Implement MLOps best practices including containerization (Docker/Kubernetes), infrastructure-as-code (Terraform), and automated deployment pipelines on GCP (Vertex AI, GKE, Cloud Run) or AWS (SageMaker).

• Collaborate with data engineers to design distributed data pipelines (Spark, Airflow, DuckDB) that feed production AI systems at scale.

• Partner with product and business teams to translate healthcare use cases into scalable AI solutions that deliver measurable ROI.

• Mentor junior engineers, contribute to GenAI/MLOps standards, and drive team-wide adoption of scalable AI practices.

• Conduct model evaluation and interpretability analysis (SHAP, LIME, RAGAS) to ensure reliability, fairness, and compliance of deployed models.

Required Qualification

• 3+ years of software engineering experience with at least 1 year focused on production Generative AI or LLM systems.

• Hands-on experience with agentic AI frameworks: LangGraph, LangChain, CrewAI, or ADK.

• Demonstrated expertise building and evaluating RAG systems (hybrid search, chunking strategies, eval pipelines).

• Proficiency with cloud ML platforms: GCP Vertex AI, AWS SageMaker, or equivalent.

• Strong Python skills; experience with ML libraries (Transformers, fine-tuning with LoRA, multimodal models).

• Experience with MLOps tooling: CI/CD pipelines, Docker, Kubernetes, Terraform, and observability stacks.

• Familiarity with data engineering tools: Spark, Airflow, SQL, distributed pipelines.

• Understanding of healthcare data privacy requirements (HIPAA) and secure AI deployment practices.

Preferred Qualification

• M.S. or Ph.D. in Computer Science, Data Science, or a related field-or equivalent relevant work experience.

• Experience with LLM evaluation frameworks: RAGAS, LangFuse, prompt regression testing.

• Familiarity with fine-tuning techniques (LoRA, PEFT) and multimodal model architectures.

• Prior work in healthcare AI, pharma, or other regulated industries.

• Published research or open-source contributions in AI/ML.

• Experience with DuckDB, structured reasoning over tabular data, or business intelligence automation.

Anticipated Weekly Hours

Time Type

Pay R