Staff Software Development Engineer - Tech Lead (Agentic)

Job description

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 We are seeking a hands-on Tech Lead / Staff Software Development Engineer to join a high-energy team driving AI transformation across our product platform. This role will be involved throughout the entire development lifecycle — designing and building AI-powered application capabilities that empower engineering teams to develop, deploy, and operate agentic AI applications at scale. The role spans the full stack : agent orchestration and backend services in Python and Java , APIs, and front-end interfaces in Angular , using LangGraph for agent orchestration across cloud-native infrastructure. You'll be equally comfortable in the agent logic, the API layer connecting it to the product, and the React/TypeScript frontend that renders it — and you'll set the technical direction for how the rest of the team builds all three. This role requires the ability to deliver efficient, production-grade solutions while identifying and mitigating risk before deployment. The ideal candidate brings strong full-stack engineering fundamentals, a track record of technical leadership and peer mentorship, and a genuine interest in using agentic AI to solve real product problems — not just prototype them. What You'll Do • Architect, d esign and build full-stack AI application capabilities — spanning agent backend services, APIs, and front-end interfaces — that enable engineering teams to ship agentic features reliably. • Design and build full-stack AI application capabilities — spanning agent backend services, APIs, and front-end interfaces — that enable engineering teams to ship agentic features reliably. • Lead technical design for agent-powered product features using CopilotKit for in-app copilots and generative UI, and Deep Agents -style architectures (planning, sub-agent delegation, long-horizon task execution) for complex workflows. • Instrument, trace, and evaluate agent behavior in production using LangSmith , driving down latency, cost, and error rate through systematic evaluation rather than guesswork. • Own the connective tissue between backend agent logic and frontend experience — API design, state synchronization, and human-in-the-loop approval flows. • Review and influence designs produced across the team; set and enforce engineering standards for agent reliability, prompt/version management, and rollback strategies. • Mentor senior and mid-level engineers on agentic systems specifically — most engineers haven't built these before; you have. • Drive platform scalability, performance, and reliability as usage grows — cost controls, concurrency, multi-tenant isolation, and infra choices that hold up under real production load. Experience with building, deploying and working hands on with Kubernetes environments • Support production issues across the full stack — frontend, backend, infra, and agent behavior alike: investigate, root-cause, and close the loop with tests, evals, and monitoring so the same failure doesn't recur. • Knowledge of tools such as Splunk, AppDynamics, LangSmith , Arize . • Build and maintain modern web technologies across the product: performant, accessible React/TypeScript interfaces, well-designed REST/ GraphQL APIs, and the infrastructure that serves them at scale. • Partner directly with product and design to translate ambiguous "make the agent do X" requests into scoped, testable technical plans. Required Qualifications • 8+ years of professional software engineering experience, with demonstrated full-stack ownership (backend services, APIs, and frontend interfaces). Must have taken several Web / agentic applications to production • 2+ years hands-on experience building and operating LLM/agent-powered applications in production — real users, real load, real failure modes, not just a prototype. • Hands-on experience with CopilotKit (or comparable in-app agent UI frameworks) — shared state, generative UI, human-in-the-loop patterns, frontend action/tool wiring. • Experience building or operating Deep Agents -style systems: multi-step planning, sub-agent orchestration, long-running or asynchronous agent tasks. • Practical experience with LangSmith (or equivalent) for tracing, evaluation, and debugging LLM/agent behavior in production. • Hands-on experience with LangGraph for agent orchestration — building, debugging, and scaling multi-step agent graphs in production. • Strong, current web technologies experience — React , modern component architecture, browser performance, and accessibility — paired with backend skills in Java and Python , and experience designing APIs that connect the two layers cleanly. • Demonstrated technical leadership: design reviews, mentoring, and being the escalation point when things break in production. • Experience working on a platform team — building capabilities and tooling that other engineering teams build on top of, not just single-product feature work. • Proven ability to manage and context-switch across multiple concurrent projects and stakeholders without dropping the details. • Strong communication skills, written and verbal — this role requires translating technical tradeoffs for engineers, product, and leadership alike, and communication is treated as a core competency, not a soft nice-to-have. • Out-of-the-box thinker who can move quickly — comfortable making sound calls with incomplete information and iterating rather than waiting for the perfect plan. Solid grasp of LLM fundamentals: prompt engineering, RAG, tool/function calling, context management, and the failure modes specific t