Battle of the vibe coders: Figma Make vs Lovable

The biggest vibe coders & AI builders are quickly becoming design tools as designers adopt them into the process of wireframing, prototyping, and early-stage testing.

Now the question becomes: "Which AI design tool should I use?"

Well, we pitted Figma Make against Lovable by using the same app idea (a gamified tarot reading app), different prompts, and got some crazy mixed results.

We dive into all of it in our second episode of UX Outpost 👇

AI tells us what to design, then beats us at our own game

Key Takeaways

  • Lovable stylistically hit the mark without needing to spell it out much (​just simple style suggestions​). It also pulled in some classic ux patterns for gamified apps.

  • Figma Make felt more like a first draft or wireframe especially since Nolan didn't add a visual reference to his prompt. It also added gamified ux patterns, but was a little more shallow.

  • Don't sleep on brainstorming concepts with ChatGPT or Claude. It asks questions you might not have thought through.

  • Deciding on a tool is all about practicing with them. So check out Nolan's free resource (​70rad.com​) to generate some app concepts (​like we did​).

Nolan's Thoughts

I'm honestly a Figma fanboy so take what I say with a grain of salt. That said, I was pretty bummed when Figma Make got outperformed by Lovable. Though, things might have improved since they upgraded to Claude 4.0 (after filming episode 2).

Here's the thing, though - both Lovable and Figma Make are basically just wrappers around whatever LLM they're using. So it's really about which AI model they pick and what extra features they layer on top that plug into our workflows.

As designers, we're already living in Figma with our design systems and libraries. Sure, Lovable's screenshot-to-code is cool, but Figma Make's integration is already more natural and only going to get better as they keep building.

Bottom line: As these tools evolve, keep an eye on what LLMs they're leveraging, and what features are added on top that lets us pass more context to those LLMs.

Andy's Thoughts

I finally got access to Figma Make since this episode aired a month ago, and it's completely changed my workflow. You can see the difference in my approach compared to the first episode...

I immediately opened a Google Doc and started with Felix's simplified recommendations from Lovable. I almost always include his "look and feel" section with minimal changes, focusing on award-winning designer language and smooth interactions.

What's striking is how much Figma Make improved between when this aired and just a week later... I literally switched from saying I'd keep using Lovable to adopting Figma Make.

The simplified prompt approach consistently produces better results than overly detailed specifications.

Get emails like these in your inbox

No spam, ever. Unsubscribe anytime.

Rad Collab | 2024, All Rights Reserved |