Working Surfacelink · 20 Nov 2026
02
On a workshop bench, a figure seen from behind turns the knobs of a small wooden test rig that holds one door handle in many sizes at once, while a heap of earlier rigs lies tossed in a crate beside the bench, the whole scene drawn with wide empty margins on every side.
LukeW

Design by Playground

Luke Wroblewski · 19 November 2026

How-to · prototyping · artifacts · Changing roles

Read the original
Takeaway

Luke Wroblewski, a product designer, argues that AI coding agents let designers build small, disposable design tools for single questions, replacing static mockups for many decisions.

Summary

Luke Wroblewski is a product designer and author who writes about interface design. On 19 November he described designers using AI coding agents to build playgrounds: small tools with controls that show a design's variants and states on real content. He shared one he built to test how much of an AI agent's activity an app should show its users.

Key points
  • A playground lets a designer change variables such as text length, emphasis, timing and loading, empty and error states, and see the result at once.
  • Building one used to take more time than it saved, so designers kept to static mockups.
  • With an AI coding agent, a playground takes minutes, and the designer can discard it once the question is answered.
  • A playground runs on the real platform, so font rendering, scrolling and screen sizes are accurate, and the chosen values can be exported as code or design tokens.
Implication

A design team can answer a single interaction question with a throwaway tool built on the target platform, instead of a set of static frames. The values it settles on can go straight to engineers or their coding agents.

Plays
Design

Build a small throwaway tool with controls for one design question, try the design's variants on real content, and export the chosen values as code or design tokens.

Derived by Working Surface from the article; more in the Playbook. Source line: Design tools used to be something you spent years learning. Now they can be something you make, use for an afternoon, and happily move on.

Source issue

20 November 2026: as AI agents make prototypes, design tools and code cheap, the scarce work becomes a person's judgement and a named person answering for the result.