How Zalando built its own AI prototyping environment on top of its design system
Sil Bormüller · 31 October 2026
Read the originalZalando's design system team built a prototyping tool on Claude Code, then an engineer's complaint about AI-written motion specifications showed that faster prototypes had not made collaboration easier.
At a Berlin meetup hosted by Into Design Systems, Dana Baddad and Gonzalo Vásquez Correa, senior product designers at the online fashion retailer Zalando, described their team's prototyping tool. Figma Make, Figma's AI prototyping feature, did not produce on-brand results, so the team built the ZDS Capsule with Claude Code, Anthropic's AI coding tool. The ZDS Capsule turns a Figma design or a prompt into a working prototype with real product data.
- The team moved its design tokens, the stored values for colours and type, out of Figma. It also split its guidelines into many small Markdown files, so that AI agents read only what they need.
- Claude Code skills, the written instructions the tool follows, gave different results from run to run, so the team has them run scripts at the critical steps.
- More than 20 designers had contributed and the team had merged 98 of their changes, although many workshop participants first got stuck on access requests, VPN, the terminal and Git.
- An engineer said the useful parts of Claude's motion specifications were tiny and obscured, and Baddad said implementation could take two or three times as long.
A design system team that gives designers an AI prototyping tool can ask engineers what they need from a prototype before generating specifications for them. That advice comes from Sil Bormüller, the founder of Into Design Systems, who wrote the account. Bormüller's page promotes a paid conference, and the Zalando team says it has no answer yet for the handoff to engineering.
Split design system guidelines into many small Markdown files with an index, so that an AI agent reads only the file it needs.
Book small-group sessions to set up access, VPN, the terminal and Git before a workshop that teaches designers to prototype with AI tools.
Derived by Working Surface from the article; more in the Playbook. Source line: AI makes exploration easy but it does not automatically make collaboration easy.
2 November 2026: as AI tools speed up design, engineering and writing, teams name a person who must be able to explain and defend the result, because the tools' own reports and specifications cannot stand in for that person.