Launch HN: A0.dev (YC W25) – React Native App Generator

105 comments

·February 11, 2025

Hi HN — we’re Seth and Ayo and we’re building a0.dev (https://a0.dev). a0.dev is a platform built to cut down React Native development time from weeks to just a few hours.

We’ve been building mobile apps together for seven years and have had several successes. One thing that’s always bothered us though is how much harder it is to build a mobile app than a website. If you’ve built an app with React Native before you’re familiar with the pains of working with both Xcode and Android Studio, writing tons of boilerplate before you can even start making the app, setting up state management, and of course going through the dreaded app review. We decided to build a platform that would make the app development process faster.

We’ve seen the success of new code-gen platforms like v0 and wanted something for React Native that goes further. We built an AI app generator that takes a user's prompt and creates a custom React Native app with an instant live preview.

Here’s a 5min demo where we recreated the Hacker News UI: https://youtu.be/f3lzBRBUous

a0.dev is great for quickly prototyping components and screens in React Native and users are able to copy the code from our generator into their preferred development environment. Our landing page has a couple of example prompts, but we encourage you to get creative when trying it out. We’ve had success generating not only functional screens like an Instagram Feed but also 2d games like Minesweeper or Flappy Bird.

Our chat has a “UI Expert” and “Advanced Logic” model users can switch between depending on the task at hand. Users can upgrade from working on a single screen to creating a full app by clicking on the “Need a Full App” button in the top right of the page. This changes the scope from a standalone chat with a single file to a full project that can include multiple chats and files. We launched an IOS app that users can download in order to preview the app on a physical device. We find that many apps look and feel better on a physical device so we recommend trying it out.

Our goal is to continue to improve the app generator while adding more features to help developers get their apps to the app store and make money from those apps. The main features on our roadmap right now are a Supabase integration and a “one click submit” button to let developers publish their app to the App Store.

There are a few limitations to note. We’re working on releasing our Android app, but Android users should be able to preview their app using the Expo Go App. The app is running React Native Web in the browser so any dependencies that don’t support web won’t work with the web preview but should work on the phone. There are also some dependencies that our system can’t handle because they require native modules that aren’t packaged into our app currently.

We hope you guys will check it out and try making an app with a0.dev. We’re available on Discord around the clock to help developers with any problems they may face and want to guide people to actually releasing their app on the App Store. Let us know what features you’d like to see and any problems you’ve faced building apps, we’d love to hear about your experience.

Here’s the link again to check it out: (https://a0.dev)

We dropped the need to sign up for the first message so you can just jump in and try it out.

Looking forward to your thoughts!

Visit

adi4213

Hey Seth and Ayo - congrats on this launch. While I think there’s certainly room for improvement, which I’ll address below - this demo makes me feel that you both can keep grinding and succeed with this. The main feedback I have is similar to others who asked how this is better than Cursor running with a react native codebase and iOS simulator running on the side. I read your response about being better at iOS/android specific things could be the case (for example, making it really nail payments and subscriptions could be useful) - but it is definitely going to be a tough battle. If someone can only use your platform and reliably have a react native codebase (version controlled on GitHub), with working Supabase auth, and fully synced with Expo (so basically almost ready to go)- that would be pretty cool! I also think there’s room for you guys to add value around some of the other annoying parts of launching a native app. If your app was able to automatically make App Store creatives and essentially publish the app for me with some user-promoting - that would be sick

sgtwompwomp

I hard agree. Seth, Ayo, I think you guys should be honest with yourself and ask whether you want to go toe to toe with Cursor, Windsurf, Microsoft, etc.

If not, take the other route. Go deep into the vertical of React Native. Help people with no experience run an ENTIRE BUSINESS just with your chatbot as the AI cofounder - you build the entire app and backend, handle marketing, publish it, all with AI agents. How sick would that be.

sethburster

Yeah, we definitely don't want to position ourselves as an IDE or pure code-gen product long term. What you mentioned with the React Native vertical really speaks to us, we've been indie app developers for a while and think there's lots of opportunities to bring AI to that field that go beyond the code.

sethburster

Thanks for the feedback, glad to hear your optimistic about the product. The code-gen field is quite competitive and we really want to work on that higher layer of app development that goes beyond the coding. What you described with the App Store creatives is something me and Ayo have talked about and would love to implement. I hope I can come back to this thread soon and let you know its live.

preaching5271

How is this different from just using Cursor on my react-native codebase? Cursor in agent mode means: no copy paste, implements features and fixes errors iteratively, knows my codebase and adheres to existing patterns. If I need changes, I just ask Cursor and preview live app changes locally

justchad

Cursor is quite a bit better as well at not adding erroneous code. Was testing a0 and it kept getting stuck in endless loops of errors. I'm sure they'll improve but it's pretty rough right now.

sethburster

We know Cursor is very good at code-gen and are improving our system to catch up. Our goal is to make the entire app development process faster and easier which involves alot of stuff outside of code-gen that we're working on and Cursor likely won't.

boole1854

> which involves alot of stuff outside of code-gen that we're working on

Could you elaborate on what extra stuff you are working on that will be a value-add over standalone Cursor?

richardw

I would guess Replit-like (or Vercel-like) automation of aspects that are RN-specific. Eg typical mobile integrations, databases, more focus on UI (design screens with drag and drop + LLM chat), secrets. Focusing means they’re able to make features work better for just one type of developer, and can move faster. Cursor would need to solve for all types of development since their users probably do everything in all languages from web to mobile to backend etc.

sethburster

Yeah, integrations like Supabase, In-App Purchases/Subscriptions setup, Notifications, and App Store submission

petrbela

Cursor is for programmers. Ultimately, you decide what the code should be.

v0/lovable/a0 will replace drag-and-drop "no-code" tools for non-programmers who don't care about code and only decide what the product should do. The tool will likely also manage hosting, either directly or through service providers, to ensure a seamless e2e experience, automatically fix runtime issues etc.

The question remains, will there be a need for Cursor and programmers in the future at all.

teaearlgraycold

Far enough into the future - of course not.

But keep in mind any point where programmers are obsolete is also the point where any job that can be done from a computer is obsolete. Including WYSIWYGing an app.

tuananh

This is good. I really didn't think of that aspect.

chaosprint

Your title is honest, and your description of the limitations is honest. It is also pragmatic to allow you to try it without registering. This is the way to success.

sethburster

Thankyou!

rushingcreek

This is an awesome demo and I hope you succeed.

I've dabbled in building a tool of this type and something I've learned is that making single-task demo apps (like chess or snake game) is much faster and simpler than making production-grade apps that could make a real business. In fact, the process of making these demo apps production ready (polishing the UI, adding features, etc.) would usually result in the user giving up. In our implementation, it was much trickier for the system to understand the nuances of polishing than getting the broad strokes of the initial demo. I'd love to know if you've ran into this and what your thoughts are on getting around it.

sethburster

Thank you for the support! Getting a polished production app can definitely be challenging if you're using just the AI. Depending on the quality bar you're going for we expect the user may need to do some coding themselves if they're struggling with the LLM. We think there are alot of apps that can be made and can make money with just the LLM today and expect to have users who are just using the LLM and also those who are willing to code themselves. As we build out more non code-gen features our offering should get more appealing to that second group of users.

cchance

I did it from the webpage, got something cool tried to add a card, and it had me signup, did google, but then my project was gone! That sucks ... losing the project during signup process is not nice UX hope you can fix that.

jmpavlec

I had the same experience, hitting back a couple times and refreshing the page allowed me to continue... But I imagine this would be a steep dropoff in engagement if you aren't taken back to the same page.

gardenhedge

Is there anything stopping you, a0, from just seeing the best apps users make and then taking that idea and launching it yourself?

ripped_britches

Only the massive amount of effort it takes to start a company

vessenes

This is a great idea. LLM workflows are not nailing native apps right now; given Apple’s slow movement, it seems like you might have a little bit of a window here.

Couple comments - have you played with Aider’s architecture mode? I think your workflows would benefit from it.

Right now it looks like the UI expert specs something and builds it, and then it asks for follow up.

I think the right dev flow working with a product dreamer involves UI mockups, then follow up questions, then implementation, with checkins — combining the idea of the architect with a “Product Manager” role is what most people need to get an app out.

Second thought is that I think you could see value with a bunch of custom prompted flows later in the app process: “do you want to take stripe payments?” “Do you want to add referral tracking support?” I’m not a native app dev, but I imagine there’s a bunch of CI, Firebase, other integrations that are best practice type things. Automating this as well would be really useful, and provide value and some lock in for your customers. To expand on this, you might make the basic option use your API access to these providers, and upcharge them. They could always implement their own if they want to move away, but the default path would get you some billing off every app that grows.

Another random idea would be ‘appifying’ a web site as a simple flow for the user. This feels like it might be even faster for the average person as a way to explain what they want, and would feed a lot of technical and visual direction straight into your workflow at the same time.

Anyway, good luck! I hope you guys get traction.

erickerr

Opening this on desktop.. I'm not reading any of this.

Glancing in more I see react-native mentioned, and expo+expo go mentioned, but like... it's been proven time and time again that Hacker News isn't your ideal feedback loop (ie: Dropbox, AirBNB).

Why make 4 paragraphs of text when you could just say:

- We obliterated X problem: (link)

- We demonstrate why that problem is significant here: (public link or private deck)

- We are growing at X->(timeframe or whatever impressive metric)

Reach out if interested. Thanks. [contact@info.com]

yard2010

Personally, I prefer the text and most importantly the story. Your textbook solution is a great method IMO, but there's less fun in this.

philjw

Looks like a great addition to v0. Great to get a headstart and do quick iterations. My process at the moment is using v0 to get a functional prototype, then import the code to cursor and build and iterate it with my custom styling / language of choice / create components from scratch but instruct Cursor to "for component.vue take xyz from component.tsx) - (v0 basically limits you to React, Tailwind and shadcn)

Recently came across this article where Gergely describes those 2 camps of "bootstrappers and iterators" - and that's basically what it is: v0/a0 for bootstrapping, cursor/copilot for actual coding https://newsletter.pragmaticengineer.com/p/how-ai-will-chang...

atymic

This looks good! Unfortunately I only made it a few prompts before it introduced an issue that it couldn't fix (back button not working). Kinda broke the whole process, what model does this use on the backend? Even after 5+ tries it couldn't fix the bug.

ayomide321

Sorry about that; this is an issue with react-native-web and its likely that it works on your mobile device. We're trying to get a solution out for this!

cbhl

Personally, I'd love to see you optimize for people who start and iterate upon an app from a mobile device and preview/use the app on the same device. Since the messages are just English text, the extra presses to type curly/square braces don't constrain you on mobile. Sometimes I get writer's block at my computer in a way that I don't get on my phone.

In particular, I'd hide/disable the sign-in-with-Google pop-up on phone-size screens (it takes a third of the screen for me), and make sure that CSS/styling makes everything fit on the code and preview screens on a phone, even in mWeb.

growthwtf

So, I use a lot of AI tools. I would say that baseline for 'coding agents' today to make it usable is that you need to recover from error messages and lints automatically. Once you can do that it will be a lot easier to use.