Senior Agentic AI Platform Engineer

CVS Health•Work At Home-Connecticut, CT

1d•$101,970 - $203,940

About The Position

CVS Health is hiring a Senior Agentic AI Platform Engineer for its AI/ML Solutions Delivery organization. The role builds foundational platform capabilities that accelerate enterprise-scale adoption of agentic AI. As a full-stack platform engineer, this person designs, develops, and operationalizes reusable components, accelerators, pipelines, frameworks, and agent harnesses to help product delivery teams move from prototype to production. The role uses technologies including Google Vertex AI, ServiceNow AI, Databricks, Claude, Python, APIs, and cloud-native services to create secure, scalable, resilient, and high-performing capabilities used across AI use cases. The engineer owns the lifecycle of shared platform assets and partners with AI Solutions Architects and Product Delivery teams to align platform investments with business needs, engineering standards, and enterprise governance. The role also provides technical leadership, coaching, and engineering best-practice guidance to contractor resources and interns.

Requirements

At least 5 years building AI/ML platforms, developer platforms, shared engineering frameworks, or enterprise-scale software solutions, with increasing technical ownership and responsibility.
At least 4 years developing reusable platform capabilities, accelerators, tooling, or shared services adopted across multiple engineering teams in enterprise environments.
At least 3 years of hands-on experience building production workloads on Google Vertex AI, Azure AI, ServiceNow AI, Databricks, or similar cloud AI platforms.
At least 2 years designing and implementing RAG, agentic AI architectures, vector search solutions, and large language model integrations.
At least 2 years designing secure, scalable, highly available distributed systems and applying AI governance, privacy, resiliency, and operational reliability practices in fast-paced delivery environments.
Bachelor’s degree from an accredited university or equivalent work experience (high school diploma plus 4 years of relevant experience).

Nice To Haves

Strong full-stack and Python engineering expertise, including production-grade development, API and microservice architecture, FastAPI, Docker, pytest, CI/CD, and cloud-native application design.
Experience building agentic AI solutions using Claude or similar LLMs, including tool use, function calling, agent orchestration, Model Context Protocol (MCP), and multi-step reasoning patterns.
Knowledge of prompt, context, and retrieval engineering; chunking strategies; embedding models; vector databases such as pgvector; retrieval tuning; reranking; caching strategies; and evaluation frameworks.
Familiarity with document ingestion and parsing frameworks such as Docling or Unstructured; SQL, schema migration tooling, ETL pipelines, and bronze, silver, and gold data architectures.
Experience supporting healthcare, pharmacy, cybersecurity, AIOps, or other regulated environments; exposure to MLOps, LLMOps, Cloud SQL, Cloud Functions, and IAM; and pragmatic technical judgment in selecting AI-powered versus traditional automation approaches.

Responsibilities

Design, develop, and maintain reusable platform components, accelerators, pipelines, agent harnesses, and shared services across Google Vertex AI, ServiceNow AI, Databricks, and Claude-based ecosystems.
Build foundational enterprise AI capabilities, including RAG frameworks, vector-enabled architectures, evaluation frameworks, orchestration services, model integration patterns, and reusable engineering accelerators.
Own platform capabilities throughout their lifecycle, including architecture, versioning strategy, production hardening, performance optimization, adoption support, and continuous enhancement; make and defend technical trade-offs.
Implement secure-by-design patterns incorporating AI governance, access controls, data privacy protections, prompt injection safeguards, resiliency mechanisms, failover strategies, and graceful degradation.
Partner with AI Solutions Architects, Product Delivery teams, Architecture, Security, and Operations to drive adoption, provide onboarding and technical guidance, and communicate architectural decisions and trade-offs to engineering teams and senior leadership.
Mentor interns and contractor resources and provide technical leadership and engineering best-practice guidance.