Agents Can Mimic Good Design. Ours Know Why It Works

Salesforce's design team built a pipeline in which AI agents produce designs from any starting point and hand engineers a specification that records the reason for every decision.

Alan Weibel, a UX architect at Salesforce, describes how the company teaches AI coding agents its design standards across several design systems. A knowledge layer holds written guidance, machine-readable design-system data, skills and linters. A pipeline called Experience Operating System uses that layer to run the design process from any starting artefact, and it writes the earlier documents that were skipped.

• Each skill first works out what is being asked, and it hands work it does not own, such as scoring a screenshot, to a separate skill.
• When guidance conflicts, the AI agent follows a written order: the user's goal, verified research, documented standards, accepted past decisions, then general heuristics.
• Evidence from reviews and Slack is collected without judgement, and a person decides whether it becomes guidance, a lint rule, an example or an eval.
• The engineering handoff includes the prototype, a specification covering layout, every interface state and keyboard behaviour, accessibility notes and the full record of reasons.

This is Salesforce describing its own internal tools. The inference, which is not the author's claim, is that a team letting AI agents produce designs needs one named person. That person decides which review findings become rules for later AI agent runs.

Derived by Working Surface from the article; more in the Playbook. Source line: Handoff is usually where the “why” evaporates.

4 October 2026: when a team prototypes before it writes requirements, the written specification engineers build from becomes a deliverable produced alongside the prototype, not something left to the builder's memory.