Loading request...
Figma Make should utilize a user's existing design systems to quickly build prototypes that accurately reflect their product with minimal prompting, as the current results are disappointing when trying to add features to a Figma frame.
Figma is threatened by Lovable, Replit, V0 and other vibe coding platforms. I thought Make would be their response but it doesn't quite cut it (yet). As more and more PMs and designers start vibe coding to convey ideas and prototype, I fully expect that we'll see less and less time spent on designs. Why go through the tedious process of getting pixel-perfect designs when you can tell the AI what you want AND get a feel for it? So when they announced "Make", I got excited. Particularly because it struck me that Figma already had access to all your design systems, so it felt like they should be able quickly build prototypes that feel exactly like your existing product with minimal prompting. But when I put it to the test that's not how things played out (see screenshot). I copy/pasted a Figma frame and asked it to add some features and the results were... disappointing. On the other hand, I took a screenshot of our existing product and asked Lovable to add a feature and it crushed it in a single shot. And the speed difference was quite material as well. To be fair to Figma, I'm not doing a 1:1 comparison. I gave Figma a frame but Lovable a screenshot. When I gave Figma a screenshot like I did with Lovable, it did really well. But then what's the edge? To boot, the ultimate design system is the code in production, so it seems like the ultimate solution will need to connect to Github. Ironically, I used Figma to collate these screenshots, so it's still very much a part of my daily workflow. Figma Make is definitely a step in the right direction, going to be keeping an eye out for future iterations, I'd love to stay in one system.