Similar to others, the unevenness of prompt results makes pricing per message quite tricky; it generated a nice looking app for me in the first go, I asked for an enhancement and it spent the rest of my free messages on trying, and failing, to fix 1 TS error that came from that enhancement. As this is using one or more of the openai/anthropic/google models, I also know that it probably won't be able to actually fix that error without my (either in code or specifically telling it what to do, which is coding) explicit help and just loop until I burn through whatever plan I have that way.
I got a cheap 1 year account on replit and it suffers from this of course; first few prompts yield amazing results and then it gets stuck. This is fine for me as I can fix it myself and by now I have good feeling of which type of error the current llms will just loop forever on, but it is hard to justify the pricing model of per message because for people who cannot fix it themselves and need to vibe on it; spend 50 bucks/mo for looping over one error for a day and that's that; pay more or wait or eject and go to another; I see non devs (vibe-only devs?) doing the latter so you never retain clients; they pick the good deals every month and move from one to the other.
So far only cursor is reasonable (that I know off) because it continues after the paid messages run out with the slower requests; you can go on forever.
I’m assuming this was named after Mr. Roarke from Fantasy Island.
All of the fantasies had a dark side.
If my prediction is correct, one of you will build your app, make billions, and then lose everything dear to you in the process.
You will then go back to Rork begging to undo what has been done, and it will tell you that it’s too late.
Then you will decide to give all of it away to charity, and then win back your previous life, but now you are quiet, scarred, and humble from your experience, thanking Rork as you leave its site in fear.
Looks like they packaged Expo into VSCode-based web editor.
I just signed up, clicked the preexisting prompt "Create an Airbnb-style home screen with a search bar at the top, filters for dates and guests, and a scrollable grid of property listings. Each listing should show an image carousel, price, rating, and basic details." and was able to preview a tab in the app on my iPhone within 5 minutes.
Pretty impressive.
But here is where the work / fun starts! You would still need a developer to revise the code, push to source control, add tests, run tests, understand how to deploy to appstore.
Vibe coding is a bit like coding 'hello world'. The actual work starts now! And that can't be vide coded away.
I was looking for a tool for building a cross platform app for my established business, I was quoted £100K+ for an app + needing a full time senior mobile software engineer.
Seems over excessive, so I found Rork which basically is all I need at close to zero cost of a senior.
Not affiliated, but just thought I'd share if Rork will help others.
I don’t want to burst your bubble but be careful how much money you decide to set on fire with Rork. In vibe coding, the solution is always 1 prompt away, it can feel like “one more and everything will slot into place”, unless you are actively coding and steering you will not be able to produce and iterate on an app with AI alone.
If your app is simple enough (1-10 prompts) you might be able to pull it off but only as an MVP (maybe that’s all you need), you won’t be able to actually launch, or rather you might launch but the app will collapse under its own weight after more prompt iterations.
And yes, if you want a good app, £100K with a full time senior dev seems completely reasonable (knowing nothing more about your app). That number doesn’t shock me. If your app is very simple then that’s high but that number is not out of them question for an app (and backend?).
Similar to others, the unevenness of prompt results makes pricing per message quite tricky; it generated a nice looking app for me in the first go, I asked for an enhancement and it spent the rest of my free messages on trying, and failing, to fix 1 TS error that came from that enhancement. As this is using one or more of the openai/anthropic/google models, I also know that it probably won't be able to actually fix that error without my (either in code or specifically telling it what to do, which is coding) explicit help and just loop until I burn through whatever plan I have that way.
I got a cheap 1 year account on replit and it suffers from this of course; first few prompts yield amazing results and then it gets stuck. This is fine for me as I can fix it myself and by now I have good feeling of which type of error the current llms will just loop forever on, but it is hard to justify the pricing model of per message because for people who cannot fix it themselves and need to vibe on it; spend 50 bucks/mo for looping over one error for a day and that's that; pay more or wait or eject and go to another; I see non devs (vibe-only devs?) doing the latter so you never retain clients; they pick the good deals every month and move from one to the other.
So far only cursor is reasonable (that I know off) because it continues after the paid messages run out with the slower requests; you can go on forever.
I’m assuming this was named after Mr. Roarke from Fantasy Island.
All of the fantasies had a dark side.
If my prediction is correct, one of you will build your app, make billions, and then lose everything dear to you in the process.
You will then go back to Rork begging to undo what has been done, and it will tell you that it’s too late.
Then you will decide to give all of it away to charity, and then win back your previous life, but now you are quiet, scarred, and humble from your experience, thanking Rork as you leave its site in fear.
Now we just need AI to simulate Charlie’s Angels.
I just tried this and it seems pretty cool. I somehow had a bad opinion of mobile apps generation, but this one actually worked.
Looks like they packaged Expo into VSCode-based web editor.
I just signed up, clicked the preexisting prompt "Create an Airbnb-style home screen with a search bar at the top, filters for dates and guests, and a scrollable grid of property listings. Each listing should show an image carousel, price, rating, and basic details." and was able to preview a tab in the app on my iPhone within 5 minutes.
Pretty impressive.
But here is where the work / fun starts! You would still need a developer to revise the code, push to source control, add tests, run tests, understand how to deploy to appstore.
Vibe coding is a bit like coding 'hello world'. The actual work starts now! And that can't be vide coded away.
I agree in that code is only a part of a whole product, but AI can absolutely revise the code, add tests, run tests, and push to source control.
But why? If you’re not a dev, it’s not like you understand or care about that.
I was looking for a tool for building a cross platform app for my established business, I was quoted £100K+ for an app + needing a full time senior mobile software engineer.
Seems over excessive, so I found Rork which basically is all I need at close to zero cost of a senior.
Not affiliated, but just thought I'd share if Rork will help others.
What's the app you created and published with Rork?
I don’t want to burst your bubble but be careful how much money you decide to set on fire with Rork. In vibe coding, the solution is always 1 prompt away, it can feel like “one more and everything will slot into place”, unless you are actively coding and steering you will not be able to produce and iterate on an app with AI alone.
If your app is simple enough (1-10 prompts) you might be able to pull it off but only as an MVP (maybe that’s all you need), you won’t be able to actually launch, or rather you might launch but the app will collapse under its own weight after more prompt iterations.
And yes, if you want a good app, £100K with a full time senior dev seems completely reasonable (knowing nothing more about your app). That number doesn’t shock me. If your app is very simple then that’s high but that number is not out of them question for an app (and backend?).