V0 feedback

Feedback
1

Hello team!

I’m a technical product designer at Emitwise.

I’ve used v0 for quickly exploring the solution spectrum for many of our initiatives. v0 has been a great partner as I dip-toed into the world of using AI to accelerate product discovery, and I will forever remember my golden time with it.

Although I’m currently not using v0 anymore (happy to share more via DM), I wanted to give you some feedback as a thank you, and specially because of being close to @pawlean :heart_hands:

Pros:

  • Really liked how clean the prototypes came out, with little to no direction needed on the visual side
  • Very easy to get the prototype running on a URL that can be shared widely
  • Forking is easy and intuitive
  • v0’s interface is minimal, and there’s no feature bloat

Challenges:

  • Working on the same project for 30+ prompts sometimes gets the project on a “broken” state: even though you can see the files in the tree view, the app preview will complain about missing files. If you download a .zip file of the project, those files the app preview complains about missing are indeed missing. Even though you still see them in the tree view, and you can click on them and see the code. I’ve had to “save” at least one project by manually going through each file and copy/pasting the contents from the web editor to my local machine. Forking the project does not fix the issue either. This is a very serious issue that has led me to stop using v0. (You are welcome to peruse the projects in my account, and you’ll see at least 1, if not more, where that has happened)
  • It’s not clear to me at all how or when does the prototype visible at a URL update. Sometimes I feel like any changes to my project will continue to reflect in the URL, and sometimes it feels like I get a unique URL for each version. An instance of the mental modal not being clear, and having received mixed signals.

Hopes?

  • The system does not proactively attempt to fix runtime errors. This can lead to significant time investment compared to other platform that do automatically attempt to fix runtime errors, and that do some level of self-testing to attempt to validate the correctness of the output (and proactive self-fix when such self-testing fails).
  • Unless I’m mistaken (and the UX hasn’t succeed in correcting my mental model), the Figma import only generates an image of the frame you have linked. It doesn’t have any impact that you wouldn’t be able to achieve by generating an image of the frame yourself and pasting into the prompt (which is faster than letting v0 obtain an image of the frame via Figma’s API)

Hope this is useful, happy to answer follow up questions. Best wishes. :slight_smile:

2 Likes
2

Thanks for taking the time to share all of this feedback! I’m sharing it all with the v0 team now

2 Likes
4

Thanks for sharing your feedback, @dasilvacontin! :folded_hands: Thorough as always :smiley:

See you at the London meet-up!

5

Just want to add that I also think that v0 is an amazing product!

I’d just add that code generation time is a real slowdown during execution. Each time I prompt I have to wait for many seconds.

I’m not sure how much y’all can influence this, and I wouldn’t sacrifice model intelligence for speed, but just that speed is a big slowing factor and affects deep work that I’d like to do.

1 Like
6

Thanks for the feedback, @aliprasla! We’ll pass it onto the team.