Stop Reviewing Schemas: How Intuit Made Developers Faster by Automating Governance
summary
Intuit replaces manual reviews with a hybrid governance system that runs real-time IDE linting plus AI-powered schema analysis and semantic collision detection during schema composition. The pipeline stitches 90+ subgraphs, runs deterministic linters and cosine-similarity checks, and blocks CI/CD publishes until issues are fixed. Metrics: 50k schema validations per month and onboarding time drops from ~90+ days (2023) to days after shipping IDE plugin (Mar 2025) and AI linter (Aug 2025).
AI-written summary · about
chapters (14)
- 0:00Introduction and Intuit overview
- 1:47Why governance fails at scale
- 5:47Manual review process breakdown
- 8:17Traditional vs AI-powered linting
- 10:02CI/CD guardrails and checks
- 12:30Demo of schema composition tool
- 13:36IDE plugin and agent support
- 16:44Collision detection demo
- 18:54Quantifying onboarding improvements
- 21:44Lessons learned from automation
- 24:14Future roadmap and automation goals
- 25:18API discovery challenges
- 26:19Proactive AI schema analysis
- 27:01Q&A session
related talks
41mOptimizing Shared Types: How Intuit Streamlined Schema Evolution in a Federated Graph
29mBold Move: Automating Away Developer Friction
07mHow Intuit Uses Argo, GitOps, and AI to Boost Developer Velocity
33mSpeed Without Sacrifice: How Wayfair Transforms DevEx with AI and MCP
46mPanel: AI in the Loop - Left-Shifting Platform DX and Governance
01mLearn about Schema Design
35mThe evolution of the GraphQL Orchestrator powering Intuit Consumer Apps
29mAccelerating GenAI Innovation: Lessons From Intuit's Agents an...