Live data from Hacker News

Launch HN: Second (YC W23) – AI bots that add features to web apps

news.ycombinator.com

11–20 of 205 posts

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#12

It's not entirely clear to me why would I want a non-deterministic language model like Chat-GPT commit into my source tree. That sounds like a generator of complexity and code pollution I won't be able to control. I want to commit my list of requirements in a well specified format and then generate a binary reproducible version of the site that is 100% stable for a certain version of the tool. If I don't like that ou…

Yea I'm totally with you! In fact, when I started Second, I was resisting GPT-3, and had planned on using a pure deterministic compiler. However, through a lot of experimentation, I discovered that the OpenAI code edit APIs, when using a temperature value of 0 (deterministic) provides shockingly good and fairly consistently results. I found that I get the best results when combining my deterministic compilers with GP…

About how soon is "very soon"?

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#13

From the video, Second looks more like a UI wrapped over a boilerplate generators that have existed for quite a while. It's a pretty powerful one, and that can be very useful in itself but how is AI involved here? If it is, was it needed in the first place? I was expecting the inputs to be more along, "give me an app to manage a restaurant", but the inputs are modules (Forgot password, billing), which are generic pie…

Great question! So all of the files are generated with either compiles or GPT-3, or if the files already exist on the repo and need to be modified, the bots use GPT-3 code edits to modify the files.

$299/project/month will sound expensive if you compare it to starter kits (usually free), templates (usually Also, yea pricing is hard! It may change.

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#14

From the video, Second looks more like a UI wrapped over a boilerplate generators that have existed for quite a while. It's a pretty powerful one, and that can be very useful in itself but how is AI involved here? If it is, was it needed in the first place? I was expecting the inputs to be more along, "give me an app to manage a restaurant", but the inputs are modules (Forgot password, billing), which are generic pie…

Many people implement things like forgot password and billing once and copy/paste in the next projects if they need.

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#15

Amazing. If it can learn my style and code conventions and create features in architectural style of the current product - it is my dream and I can pay a lot for it.

Yea that's the plan! Currently I am not building fine-tuned models on top of your codebase (and potentially other related code bases) but I will be doing that soon.

I have found that even without a fine-tuned model, the OpenAI code edit and code writing APIs are actually quite good at the file level. Fine-tuned models will just make it even better!

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#16

From the video, Second looks more like a UI wrapped over a boilerplate generators that have existed for quite a while. It's a pretty powerful one, and that can be very useful in itself but how is AI involved here? If it is, was it needed in the first place? I was expecting the inputs to be more along, "give me an app to manage a restaurant", but the inputs are modules (Forgot password, billing), which are generic pie…

Great question! So all of the files are generated with either compiles or GPT-3, or if the files already exist on the repo and need to be modified, the bots use GPT-3 code edits to modify the files. $299/project/month will sound expensive if you compare it to starter kits (usually free), templates (usually Also, yea pricing is hard! It may change.

And this boilerplate can't be added fast with Copilot? Which is $100/year.

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#17
Agreed with the comments on the demo video music. It was startling.

I'm pretty skeptical still as all your demo has shown is similar to what a boilerplate tool might do, but with more unknowns. You have to start somewhere, though, and I think you're on a decent path.

I've been experimenting with chatgpt to do similar workflows to what you're describing for your vision of Second. From my experiments, it seems your demo may be approaching close to the current limits of gpt models. Fine-tune training on my existing code base and requirements may help, but I still feel like we're a very long way away from your vision for Second. Maybe even far enough away that it won't ever be realized.

Still, you've already built a much better workflow than my experiments have yielded so far, and I think it is a very exciting proposition.

Feel free to disregard, but I think I'd approach the business differently than you have - raise your prices drastically and instead of a dev-facing tool to start, make your offering more like an agency. Hire devs and designers to build the client work and sales people to find more deals. Have devs and designers use your tool, providing feedback. Restrict modes of communication to AI-parseable docs between the client and devs/designers. Collect data as you iterate through as many projects as you can find and staff to deliver. Then train/fine-tune LLMs on that whole dataset. Iterate on your tool throughout this whole process. Even better if you could embed yourself/your tool into existing agencies and dev shop firms, but that seems like it could be difficult to navigate.

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#18
Congrats! Looks useful. Can you explain the difference between Starter, Pro, and Maintenance Bots? $299/mo is pretty steep w/o a trial period. If you're against a trial period, what's the refund policy? Can we change the project it's pointing to at any given time? When does the 50% promo end?

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#19

It's not entirely clear to me why would I want a non-deterministic language model like Chat-GPT commit into my source tree. That sounds like a generator of complexity and code pollution I won't be able to control. I want to commit my list of requirements in a well specified format and then generate a binary reproducible version of the site that is 100% stable for a certain version of the tool. If I don't like that ou…

Yea I'm totally with you! In fact, when I started Second, I was resisting GPT-3, and had planned on using a pure deterministic compiler. However, through a lot of experimentation, I discovered that the OpenAI code edit APIs, when using a temperature value of 0 (deterministic) provides shockingly good and fairly consistently results. I found that I get the best results when combining my deterministic compilers with GP…

Small warning: 0 temperature still isn't deterministic with openAI endpoints. So, if you are relying on this as an absolute, it'll definitely fail at some point (this bit me at one point recently). If you just like it because it's "mostly deterministic" so it feels more well behaved, then it's fine.

From here: https://platform.openai.com/docs/guides/completion

> Setting temperature to 0 will make the outputs mostly deterministic, but a small amount of variability may remain.

Re: Launch HN: Second (YC W23) – AI bots that add features to web apps

#20

Earlier quoted context omitted.

Great question! So all of the files are generated with either compiles or GPT-3, or if the files already exist on the repo and need to be modified, the bots use GPT-3 code edits to modify the files. $299/project/month will sound expensive if you compare it to starter kits (usually free), templates (usually Also, yea pricing is hard! It may change.

And this boilerplate can't be added fast with Copilot? Which is $100/year.

Maybe. Copilot is more like an extreme programming experience, where you are still piecing together things yourself in the IDE. Second does all of that for you by building complete features and raising a pull request!

I believe that in the future, A.I. bots will be used to build large chunks of the codebase (like Second). And then human engineers will use A.I. bots in the browser to make low level tweaks and additions (Copilot)

Post reply on HN