Senior Software Engineer, Frontend Platform & Design Systems
Connor Dibble
Most of my work sits where product workflows become software. I build the design systems, service contracts, and AI tooling that keep large engineering orgs shipping.
About
In practice that means component APIs, GraphQL/REST contracts, design-token pipelines, Figma-to-code tooling, and the governance that keeps large frontend orgs aligned. At State Farm, I lead the statefarm.com Digital Experience platform across customer and agent-facing surfaces and steward SFDS, the enterprise design system used by 1000+ engineers and designers. Outside work, I build focused AI products through Proceris around R&D documentation and workflow automation for small teams.
More about meProjects
DesignRail
Open source
Human-in-the-loop design-system mapping platform for AI-assisted implementation. Translates Figma component intent into Web Component mappings, validates compliance, and records human review decisions through a GraphQL API.
- React
- TypeScript
- GraphQL
- Drizzle
- Figma MCP
SFDS
State Farm — closed source
State Farm's canonical enterprise design system and migration target for statefarm.com. I led sf-ui, a framework-agnostic Lit/Web Components library with 70+ primitives, plus the Figma Variables → W3C token pipeline and npm/CDN release governance for 1000+ engineers and designers.
- Web Components
- TypeScript
- Design Tokens
- npm + CDN
Agent Feedback Platform
State Farm — closed source
Full-stack AI feedback platform commissioned by State Farm's EVP of enterprise technology. Went from concept to production in two months for its debut at agency convention, and the pilot earned backing to become the official support and feedback channel for all 10,000+ State Farm agents.
- Angular
- TypeScript
- Node.js
- AWS
- AI Workflows
ResearchLog
Proceris — closed source
Full-stack R&D tax-credit documentation SaaS that maps real engineering work to IRS Section 41 evidence. Ingests GitHub PR and issue activity, classifies work with Claude at the pull-request level to keep model spend bounded, and keeps reviewer overrides tenant-scoped with Supabase RLS.
- Angular
- TypeScript
- NestJS
- Supabase
- Inngest
Experience
Senior Technology Engineer — Digital Experience
State Farm · Mar 2026 – Present
Set technical direction across the statefarm.com Digital Experience suite: 8 product teams, 40–50 engineers, and an executive-backed migration toward 100% SFDS adoption. Shape GraphQL contracts, service-integration patterns, and AI tooling for customer-, agent-, and internal-facing workflows, including a Figma MCP pipeline that translates fully designed pages into sf-ui Web Component implementations.
Lead Software Engineer — State Farm Design System (SFDS)
State Farm · May 2023 – Mar 2026
Led engineering for SFDS, State Farm's canonical enterprise design system. Architected sf-ui, a framework-agnostic Lit/Web Components library of 70+ primitives, and a Figma Variables → W3C design token plugin in TypeScript. Owned npm + CDN release governance for 1000+ engineers and designers.
Software Engineer — Drive Safe & Save Telematics
State Farm · May 2020 – May 2023
Built Node.js Lambda APIs and Java Spring backend services for the Drive Safe & Save telematics enrollment experience on statefarm.com. Owned frontend direction for the product suite across React apps and micro-frontends.
Skills
- Design Tooling
- Figma Plugin/API · Figma MCP · W3C Design Tokens · Design-to-Code Automation · Component Governance · Accessibility · Documentation Systems
- Frontend / Platform
- React · TypeScript · JavaScript · Lit / Web Components · Angular · SPA Architecture · State Management · Cross-Framework Integration
- Backend / Cloud
- Node.js · Java · AWS Lambda · GraphQL · REST · API Contract Design · Service Integration · DB Schema Design · Event-Driven Systems · Serverless · npm / CDN Delivery · GitLab CI/CD
- AI + Reliability
- Application-Level GenAI · Agent Workflows · MCP Servers · Agent Skills · Structured AI Outputs · GitLab Webhook Agents · SLOs & Observability · Dynatrace RUM · AWS CloudWatch · Incident Response