Inspiration

We recognized the efficiency that Github Copilot brought to development, and we wanted to do something similar for another industry. Rather than just having Copilot for development, we wanted to use the powers of Copilot for making beautiful UI designs in a fraction of the time that it would normally take.

What it does

Figma Copilot analyzes your existing UI, and generates new suggestions for you to improve your UI using AI.

How we built it

We built a Figma plugin using Typescript, which sent prompts to a GBC which in turn sends them to Stable Diffusion.

Challenges we ran into

We ran into lots of challenges with the front-end and the back-end. Typescript gave us multiple challenges when it came to using Figma's API. Also, working with docker images and detectron provided many roadblocks.

Accomplishments that we're proud of

  • Getting the Figma API to work properly.
  • Getting all the libraries to work cohesively.
  • Creating a workable flow for this idea to actually work.

What we learned

  • Don't use Figma API again.
  • Stick to JavaScript
  • Hackathons are all about the music.

What's next for Figma Copilot

  • Cleaner UI, and the ability to auto-suggest.

Built With

Share this project:

Updates