Senior Full Stack Engineer, AI Platform

Overview

Geisinger is operationalizing AI at scale, moving from pilots to a portfolio of production AI capabilities across clinical care, operations, the health plan, and pharmacy. Internal applications—dashboards, evaluation tools, and developer portals—enable program teams, governance stakeholders, and platform engineers to use and trust the AI platform. The Senior Full Stack Engineer owns the end-to-end product quality of every user-facing application the AI Platform ships, from the design system to component library, backend-for-frontend APIs to telemetry, and role-based access to load-time budgets.

This role ensures that platform dashboards, evaluation tools, and internal portals are trustworthy and compelling enough for continued use.

The AI Platform is an enabling team building reusable capabilities for every AI program t capabilities have a user-facing surface—dashboards, configuration interfaces, review tools, developer portals. Every surface is an opportunity to help or hinder adoption. The portfolio evolves with platform capabilities, and someone must own the quality bar for how these applications look and feel.

• The frontend architecture and the shared design system / component library used across every AI Platform application — visual and interaction consistency.
• RBAC-aware interfaces — ensuring program teams, governance stakeholders, leadership, and platform engineers see the right data and controls for their roles.
• The backend-for-frontend layer — API patterns, data contracts with MLOps, and the interfaces where the UI meets the platform.
• Application-level testing strategy — unit, integration, and end-to-end tests that the deployment verification pipeline triggers automatically.
• Usage telemetry and analytics — instrumenting applications for adoption, feature engagement, error rates, and load times to inform the platform roadmap.
• Frontend performance — bundle optimization, caching, lazy loading, and load-time budgets for internal applications.
• UX quality gates — defining and enforcing the standard that no application ships below.

This role operates at three altitudes:

• With the design system (hands-on build) — Own the shared component library and patterns. Build accessibility from the start; make the design system attractive for other engineers to consume because it speeds their work.
• With application teams (collaborative delivery) — Build user-facing applications with the MLOps backend, negotiate API contracts, RBAC implementations, and data shapes. Today’s portfolio is one set of applications; tomorrow’s will be another, and you’ll own the interface layer through transitions.
• With the data the applications generate (product instinct) — Wire telemetry into every surface, read it, and bring evidence to roadmap discussions. Use instrumentation to drive future investments and retire less-used surfaces.

The platform team maintains a growing set of internal applications, including:

• An AI Scorecard dashboard displaying per-initiative status across Performance, Adoption, Outcome, and Equity pillars
• LLM-as-Judge evaluation tool for configuring evaluation criteria, reviewing judge outputs, and tracking eval results
• Developer portals and internal tooling — documentation sites, onboarding flows, and self-service interfaces for the platform’s standard delivery path

Expect the portfolio to evolve as new platform capabilities come online. The role is the application layer of the AI Platform, not these specific applications.

• React 18 (frontend framework)
• Mantine and/or MUI (design system foundations)
• FastAPI (backend-for-frontend APIs)
• SQL Alchemy (ORM / database access)
• Docker (containerization)
• Jest, React Testing Library, Playwright or Cypress (testing)
• Open Telemetry and application-level metrics (telemetry and observability)

• MLOps Engineers — consume their APIs for the Scorecard and Eval Tool; align on data contracts, feature requirements, and API versioning.
• Sr. Platform Engineer — coordinate on deployment targets, CDN, environment configuration, IDP self-service provisioning, and…