Staff Software Engineer, Agentic AI

About Atria Atria is a membership-based preventive health care practice delivering cutting-edge primary and specialty care from the comfort of your home, at our practices in Palm Beach and New York, or wherever you are in the world. We bring together a multidisciplinary team of renowned, in-house physicians to provide proactive, preventive, and precision-based care for Atria members and their families. We aim to optimize the lifespan and healthspan of all our members through meticulous screening and tailored interventions to prevent, reverse, or manage all major chronic diseases. Each member’s care is led by a dedicated Chief Medical Officer who collaborates on your behalf with specialists in cardiology, neurology, pediatrics, gynecology, endocrinology, performance and movement, and more. Our exceptional clinicians also work closely with the 60+ members of the Atria Academy of Science & Medicine, top experts in their respective fields who are available for rapid consults, support, and referrals. Role Overview Atria Health is seeking a Staff Software Engineer, Agentic AI to define the strategy and lead the implementation of emerging technologies across our Technology organization and Atria as a whole This is a senior individual contributor role for a technical leader and builder who is equally comfortable shaping strategy, writing production code, designing governance, and driving adoption. You will identify the highest-leverage opportunities to apply agentic AI across Product, Design, and Engineering; architect and ship the systems that make those workflows faster and better; and establish the measurement, evaluation, and security standards that let us scale confidently. This role requires someone who actively works at the frontier of AI-assisted development . You should already be using modern AI tooling and agentic workflows as part of your daily work and continuously seeking to leverage the latest developments. This is not a pure strategy role. We are looking for a seasoned engineer who will build dependable systems, drive their adoption, and show measurable impact. Your initial focus will be applying agentic AI across Atria’s technology workflows . Working with engineering, product, design, and cross-functional partners, you’ll identify high-leverage opportunities, define where human oversight is required, and establish clear metrics proving improvements in speed, quality, and reliability. After establishing successful systems within the Technology organization, you will explore and drive high-impact applications of agentic AI across other parts of Atria. Tech Stack Languages: TypeScript Backend: Node, Express Frontend: React Infrastructure: Google Cloud Platform, Terraform Database: MySQL Tooling & Integrations: Cursor, Github, Sentry, Datadog, Mixpanel, Omni Key Responsibilities Strategy & Opportunity Selection Identify the highest-leverage opportunities for agentic systems and emerging technologies across Product, Design, and Engineering workflows Map end-to-end workflows and turn ambiguous opportunities into clear technical strategies and phased execution plans Prioritize work based on measurable business value, technical feasibility, operational risk, and adoption potential Help Atria stay current on meaningful advances in emerging technologies and translate them into practical plans Platform & Workflow Development Architect and build Atria’s internal platforms for agentic workflows, including context retrieval, integrations, permissions, validation, observability, and auditability Design reusable services, APIs, patterns, and interfaces that support AI-assisted workflows across planning, implementation, review, release, and incident response Build and ship high-impact workflow improvements yourself, especially in the highest-priority and most technically challenging areas Create systems that can safely combine context across code, documentation, design files, collaboration tools, analytics, and other internal sources Define integration patterns that allow new workflows to be added without rebuilding core primitives every time Evaluation, Reliability & Governance Establish evaluation frameworks for agentic systems Define how output quality is measured and how systems are monitored over time for drift, regressions, and operational failures Instrument workflows with meaningful metrics tied to adoption, cycle time, review efficiency, quality, reliability, and user trust Partner with Foundations Engineering and other stakeholders to implement appropriate controls around permissions, access, audit trails, vendor risk, and sensitive data handling Define where agents can act autonomously, where human approval is required, and how those boundaries are enforced in tooling Enablement & Adoption Create playbooks, standards, and training that help engineers, product managers, and designers adopt effective AI-assisted ways of working Raise the bar on internal practices around context management, prompt and work