Principal Architect - AI Builder

CVS Health•Work At Home-Illinois, IL

2d•$144,200 - $288,400•Remote

About The Position

We are hiring a Principal Applied AI Architect to take ownership of how we build AI‑powered products and applications at CVS Health. This role sits within our Enterprise Architecture organization and is responsible for designing, building, and shipping AI applications across the enterprise. LLMs, agents, and AI coding tools have matured to where they can drive real production systems, and we are investing in this capability across the enterprise. We need someone who can identify the right products to build, design and architect each system, write the specs, and ship working applications using modern AI coding agents while keeping production quality high. We use foundation models from Anthropic, OpenAI, Google, AWS Bedrock, and Azure OpenAI, and build agentic systems on frameworks such as LangGraph, CrewAI, OpenAI Agents SDK, and MCP. AI applications must integrate cleanly with our enterprise platforms, identity, and data services, working alongside the platform and system owners who run them. This role also includes coaching other architects and senior engineers on advanced applied‑AI coding techniques and AI‑augmented workflows so the existing team can move faster on their current work. This is a hands‑on role: you will write PRDs, design systems, build prototypes that become production references, ship code using AI coding agents, review AI‑generated code, coach other architects, and present to leadership.

Requirements

12+ years in software engineering and architecture, with a demonstrated progression from production developer to enterprise architect to applied‑AI builder.
8+ years operating at the architecture altitude with a track record of systems you designed, built, and shipped.
2+ years building and shipping production AI applications using LLMs, agentic systems, RAG, MCP, and evaluation frameworks.
Hands‑on experience with at least one major foundation model provider (Anthropic, OpenAI, Google, AWS Bedrock, or Azure OpenAI) and at least one agent framework (LangGraph, CrewAI, OpenAI Agents SDK, Semantic Kernel/MAF), with a clear opinion on tradeoffs.
Strong Python plus one additional production language (TypeScript, Java, Go, or similar).
Solid system‑design fundamentals across distributed systems, APIs, cloud (Azure, AWS, or GCP), data pipelines, and enterprise integration patterns.
Strong product instinct: able to identify what to build, scope it, and write the spec independently.
Proven track record of AI applications shipped to production with disciplined evaluation, testing, and observability.
Able to communicate clearly with both engineers and senior leadership.
Bachelor’s degree in Computer Science, Engineering, or a related field, or equivalent work experience.

Nice To Haves

Experience working with multiple foundation model providers and a clear point of view on when to use each.
Experience building MCP servers and defining tool integrations for agent ecosystems.
AI evaluation tooling experience (LangSmith, Braintrust, Promptfoo, or custom evaluation harnesses).
Hands‑on experience using AI coding agents (Claude Code, Codex, Cursor, GitHub Copilot) as production build tools, not just IDE assistants.
Experience publishing PRDs, product specs, or technical strategy documents as a hands‑on architect or staff engineer.
Contributions to open‑source AI projects or communities (LangChain, MCP, agent frameworks, etc.).
Healthcare, insurance, or financial services experience (HIPAA familiarity).
Patent filings, conference talks, or published technical writing on applied AI.
Cloud certifications are a plus but not required.

Responsibilities

Identify high‑value AI use cases across CVS Health and shape them into products with clear scope, user need, and success criteria.
Design and architect each AI system end‑to‑end, including retrieval, agent orchestration, tool and MCP integration, evaluation harnesses, observability, and human‑in‑the‑loop hand‑offs.
Write product requirements, design documents, and architecture decision records for the AI products you own.
Build production implementations using AI coding agents (Claude Code, Codex, and equivalents), with disciplined testing, review, and observability.
Partner with platform and system owners to integrate AI applications cleanly with our enterprise platforms, identity, and data services, following CVS standards for security and compliance.
Use modern AI coding agents (Claude Code, Codex, Cursor, and equivalents) as a primary production build tool to compress build cycles while maintaining high production quality.
Apply seasoned engineering judgment to review AI‑generated code at architectural depth, catching integration mistakes, silent failure modes, and design drift before they reach production.
Coach other architects and senior engineers on applied‑AI coding techniques and AI‑augmented workflows so they can speed up their own work and their teams’ work.
Build reusable building blocks (agent skills, MCP servers, sub‑agents, prompt and evaluation libraries, reference implementations) from the products you ship so other teams can adopt them.
Identify where coding agents and agentic workflows can speed up application, platform, and integration team work, and help those teams adopt the practice.
Participate in cross‑functional architecture working groups covering applications, platforms, and integrations that use AI.
Publish architecture decision records and reference implementations from the products you build so teams can adopt them.
Mentor architects and engineers and conduct architecture reviews of AI systems built across the organization.
Work with security and compliance on HIPAA, SOX, and PCI requirements as they apply to AI systems and data handling.
Represent CVS Health in vendor evaluations and partner programs (Anthropic, OpenAI, Google, Microsoft, Databricks) and stay connected to the open‑source AI ecosystem.