Teaching agents product design at Vercel - Vercel
vercel.com · 3,875 words · saved by 2 readers
Learn how Vercel teaches agents product design with agent skills, lint rules, Vercel Agent code reviews, evals, and a human-led update loop.
Coding agents can produce working UI fast, but what's harder is a different shape. They can copy your product's style, match its patterns, and try to follow its conventions. What they cannot do is understand why those patterns exist. Code shows agents what shipped, not why one component, phrase, or interaction became your standard. That reasoning lives in design reviews, PR comments, Slack threads, and with the people who were in the room. For an agent, context that isn't in the codebase doesn't exist. Vercel is an agent-native team. We treat accepted product decisions like code, keeping them
saved by
related reading
- Using AI as a Design Engineerjakub.kr
- How our agents build on-brand pages with design.mdvercel.com
- UI Skills for Design Engineers | UI Skillsui-skills.com
- Design Engineeringfloguo.com
- Design Engineering at Vercel: What we do and how we do it - Vercelvercel.com
- Demystifying evals for AI agents \ Anthropicanthropic.com
- AI agent evaluation frameworks for production - Vercelvercel.com
- Why We Built Our Own Background Agentbuilders.ramp.com
- Impeccable: The missing upgrade to Anthropic's impeccable skillimpeccable.style
- GitHub - vercel-labs/web-interface-guidelines: Guidelines for building interfaces on the Web.github.com
- How Figma stays ahead of vulnerabilities with agentsfigma.com
- AI Design Field Guideaidesignfieldguide.com