v0: Generate UI with simple text prompts | Hacker News Reader

v0: Generate UI with simple text prompts

v0.dev

69 pointsleerob40 comments

I thought registration walls that do nothing but allow you to join a waitlist were discouraged/banned here?

That's a rule for "Show HN" posts, not posts in general. If this had been "Show HN: v0", the mods would strip the "Show HN" off when they noticed.

Is there any reason to prefer to be a “Show HN” over a general post?

It has its own, stricter rules and site section which some people also follow so in theory you get more shots at exposure. Also it's a signal the poster is the author of the thing being showhn and the purpose of the tag is feedback so you get more of that.

Downside is you have to have something to show.

You can use the prompt library (https://v0.dev/explore) and copy/paste any of those examples. We're going to be rolling out access to folks on the waitlist, but for right now it's still in private alpha.

You should make it clearer that this is a preview that people can't play with (it sends exactly the opposite message right now), because the design you have now is really begging for a threat that is 90% complaints. It's just the way HN is.

Good feedback, thank you

Fair warning: there is an registration wall in front of the service.

No, a registration wall in front of a wishlist of a service.

Flagged.

You've been braver than I was. Thank you.

This isn't a valid reason to flag a post. A good reason not to upvote it! But it being a waitlist doesn't make it off-topic for the site. It doesn't matter that much, but if you habitually flag posts for the wrong reasons, your flags will stop doing anything on the site.

Sorry, but this isn't just a waitlist page, it first lets you give your input, then asks you to register, and then adds you to the waitlist, it just doesn't work.

If it didn't use the dark pattern, I would agree with you. Ask the user to generate input, then ask them to register, then inform them they can't use the service. Flagged.

We can try to get an Official Ruling but I'm pretty sure you're wrong about this. Waitlist signups are heavily moderated, almost always right off the front page, whether they are in Show HN or not. Waitlists are highly flaggable, pretty much like pre-nouncements which get the same moderation treatment.

Tangent fight: people should flag more, not less and the chances you get caught 'misflagging' a front-page article are pretty close to nil probably because, as you say, it doesn't matter that much.

https://hn.algolia.com/?dateRange=all&page=1&prefix=true&que...

Edit: nvm, it's already in the hadith and alwayshasbeen.gif

https://news.ycombinator.com/item?id=22124736

https://news.ycombinator.com/item?id=14310713

It's not strictly a waitlist! (I'm fine either way.)

Yes but I'm rulethumping about

it being a waitlist doesn't make it off-topic for the site.

Point taken! I'll remember that going forward.

New generations are behind a waitlist, but you can see what the service does on the explore page (and bottom of homepage linked above) https://v0.dev/explore

I've been playing around with something similar. It's buggy and in beta but you can use it now with no waitlist: https://gitwit.dev/

Very excited to use this. I recently started working on rebuilding my personal website and using that learn about RSC / App Router. As I started down the path, I learned that ChakraUI (which we use at work with Pages Router) doesn't work with RSC. I knew of shadcn/ui but I have literally zero experience using Tailwind; I'm sure I'll be able to figure it out, but having something like v0 will vastly improve my speed at which I can build my website and learn from generated examples. For someone with no design chops I can't wait to start playing around.

Things like this needs one to be creative before the prompting part, creativity first, then with help of llms you can get a code,based on your creativity, only have to think and work with AI as your assistance to make logical flow of your input or prompts, what I have noticed is, the engineers are less creative,but good at coding what they have on their heads, so imagine having both imagination and coding skills and also not forgetting the know how of what you are actually trying to create, example, we can have a creative engineer, that doesn't know how to trade (financial market) how so you expert that person to come up with a good project? They need to firstly understand how to trade, risk management, position sizing, a strategy, entry and exit points, so without this knowledge, the project is good as useless

Hi everyone! Jared Palmer (https://x.com/jaredpalmer) here from the v0 team and Vercel. Happy to answer any questions. The team is very excited to finally share what we've been working on with the community. We know it's early, but as the name implies... it's v0.

Link: https://v0.dev | FAQ: https://v0.dev/faq

Looks interesting, are everyone's generations made public though? A private option would be nice.

Yes! This is something we'll be adding soon!

If you want a good response from HN, it would help to let us past the waitlist. Right now it's just a gallery for existing generations.

Can you talk at all about the tech that goes into making something like this? Or about the problems you faced building a system for production?

This is useful. Thank for you creating this.

Next step: getting AI to replace or generate the code to replace the UI elements with data.

The future of backend:

I would like software architecture to be transformable into vectors so that we can "understand" architecture and move software architecture from one design to another, all without breaking everything.

The code is just a mechanical output of semantic understanding of what is required then. The vectors are your code.

EDIT: We'll no longer be creating YAML but describing how things are to be.

This might be the biggest piece of shit I’ve seen in the web development space in the past few years. It’s truly peak Vercel.

Ok, but can you please not post like this to HN? You may not owe web development companies better, but you owe this community better if you're participating in it.

https://news.ycombinator.com/newsguidelines.html

You can't actually try it without a Vercel account. Do you need a paid plan too?

It's in private alpha currently, but we'll be opening up the waitlist in the future. It uses Vercel for OAuth only, your plan doesn't matter. In the meantime, you can check out the prompt library: https://v0.dev/explore

I'd be very interested to understand how this works, does anyone know if they've released any guides or if someone has documented trying to do something similiar?

This looks like nightmare fuel.

I honestly think front end is kind of over or has a shelf life of maybe a few more years. Not sure how to go about planning my career according to this belief

A lot of intellectual/software work is a bit like this though.

It'll be interesting to see whether these AI tools scale into more complex work, as right now what you tend to see is a "good start point" but eventually the context of what you're building gets too large and the AI starts to produce junk.

I disagree, because most of the frontend work I’ve been involved with over the past 10 years or so isn’t just blindly implementing mockups. Actual UI engineering work usually involves making the feature a logical reality. Product and UX have a great ideas, but they’re not engineers and it usually takes a lot of work to make these ideas match the reality of the domain, making things consistent with UI patterns, existing components, web standards, etc.

I’m honestly not worried about AI being the programmer for UX people, that’s for sure. I think creating mockups are a bigger target for AI.

Also, this work is usually the best starting point for a meaningful domain API, and helps drive the backend as well.

well, thats not how it will work, wait 5 more years, you will be either jobless or will have to transit to something else. Frontend is either for bots or for extremly creative people who can make stuff for bots, so they can get trained on it.

and I agree if something like this can be done, Entire software engineering is going down, Most people even in the very complex seeming tasks are doing the copy paste thing, all day

Maybe, but backend seems more squeezed at companies I've dealt with TBH. Not AI related, but startup I was at introduced Hasura (GraphQL on top of Postgres), and eliminated most of the work the backend folks were doing (even though they didn't like mapping frontend API calls to data all day)

I’m not convinced of that yet. Visual UI builders have been a thing for decades and this is more or less the same thing except the input mechanism has changed. Instead of point and click you’re prompting it with text.

I think there are a lot of problems, but mainly I just don’t think it’s ever going to be practical to write and maintain software (front or backend) using conversational text prompts. It’s a very imprecise tool for the job.

so lame

Reply on news.ycombinator.com