Design system debt used to be paid by one designer, quietly, in Figma files nobody else opened. Now it gets paid by the engineer whose AI-generated code ignores every token, spacing rule, and component boundary the design team defined.
That shift is the throughline of this week’s Design Bootcamp collection, which bundles three resources worth bookmarking:
- Design system debt in the AI era — When Cursor or v0 generates a button, it doesn’t check your design tokens. It invents padding, picks a hex color, and ships. The debt moves from Figma layers to production code, where it’s harder to spot and more expensive to fix.
- Game UX principles applied to product design — Game designers have spent decades solving problems product teams are hitting now: onboarding without manuals, progressive disclosure of complexity, feedback loops that keep people engaged.
- 39 principles for AI interaction — A reference list covering latency disclosure, failure recovery, and uncertainty communication, written specifically for interfaces where system output is non-deterministic.
The design-system-debt piece hits hardest. AI coding tools don’t read your Storybook. They don’t know your spacing scale. Every generated component is a potential debt injection point, and the person catching it is no longer a designer reviewing a Figma branch — it’s a developer mid-session who may not know the design system exists.
NOTE
The “Game Design Debt” concept (Tavrox’s definition) draws a parallel: you must build the core foundations of a genre before iterating on it. Same applies to design systems — skip the foundation, and every AI-generated layer compounds the debt.
The practical move: if your team uses AI coding tools, your design tokens need to live somewhere the tools can actually read — not just Figma, but as JSON or CSS variables committed to the repo. Otherwise the debt piles up faster than any single designer can pay it down.