Vibeleaderboard
← All Intel
Intel / article

How We Built Ai Search In Figma

Source
figma.com
Author
Figma Blog
Date

Context

Figma's AI search grew out of work on an assistant that would suggest the next design component. User research redirected the team toward an earlier obstacle: designers needed to find existing work before they could reuse it. Its October 2024 account describes prioritizing search by screenshot, selected design or text description.

Finding a visually similar result was only part of the task. The interface also added ownership and editing information, filters and previews to help designers judge relevance before opening a file. The companion infrastructure account explains how designs became searchable.

Terms in this piece · Glossary
  • embedding — A list of numbers representing a piece of text's meaning, so that similar meanings end up numerically close and can be searched.
  • RAG — Retrieval-augmented generation — fetching relevant documents first and pasting them into the model's context so it answers from your data instead of memory.
Read the source www.figma.com
More from Figma Blog
Recommended reads
Comments

Checking sign-in…

Loading comments…