This is so neat (and useful!). Thank you both for sharing it, and for thoughtfully documenting your thought processes around it.
I've recently started vibe coding and -- while I never have a lack of ideas for features and functionality! -- I just haven't yet developed a great design sense, so this is a really great starting point :)
P.S. -- One thing I'm curious about: why here.now? I've seen the site before and it seems quite generous with its free offering, but how is this better / more useful than just using free ChatGPT hosting or hosting for free via Github or Cloudflare?
i found github pages just a bit annoying to deal with to be honest - and i dont always put everything on github. cloudflare i use sometimes but always found it had so many more features than i needed. i had poor experience with chatgpt sites, being slow but i should give it another go.
i am an investor in here.now too btw! but used them first and liked so much i had to invest 😊
I've just had Claude and ChatGPT deal with Github directly; all I did was set up an account there, so I haven't had to deal with the complexity myself.
With ChatGPT Sites... the one app I've hosted there (www.CardValuator.com) seems to load pretty fast, but it's also pretty much a compact one-page sorta site :).
Perhaps for my next vibe-coding project, I'll check out here.now! I think it's cool that -- unlike Substack -- they support BYODomain without charge :).
I made a seamless pattern generator—seamlessify (https://seamlessify.com/). It eliminates image edge seams to create infinitely tileable patterns (also supports AI generation from prompts).
This is a great idea - design vocab is exactly where I get stuck prompting agents. One related gap I keep hitting: once the designs are generated, I often need to merge two images (a layout over a background, before/after pairs). I've been using https://aiimagecombiner.app for quick composites instead of opening a full editor.
Ben, re: “I kicked this off in Pi using Fable 5.1 - I just wanted to brainstorm and play with a few ideas first, when I wanted to build it for real I used Factory’s Droid.”
its super context light and gets out of the way - so to have the clearest ‘mind’ i use pi. with other coding agents their system prompt gears them up for coding tasks.
its like sometimes if i need a less agent-y agent i’ll just ask plain http://claude.ai or http://chatgpt.com sometimes even with icognito to make sure no context is misleading it
This is so neat (and useful!). Thank you both for sharing it, and for thoughtfully documenting your thought processes around it.
I've recently started vibe coding and -- while I never have a lack of ideas for features and functionality! -- I just haven't yet developed a great design sense, so this is a really great starting point :)
P.S. -- One thing I'm curious about: why here.now? I've seen the site before and it seems quite generous with its free offering, but how is this better / more useful than just using free ChatGPT hosting or hosting for free via Github or Cloudflare?
i found github pages just a bit annoying to deal with to be honest - and i dont always put everything on github. cloudflare i use sometimes but always found it had so many more features than i needed. i had poor experience with chatgpt sites, being slow but i should give it another go.
i am an investor in here.now too btw! but used them first and liked so much i had to invest 😊
Ah, good to know!
I've just had Claude and ChatGPT deal with Github directly; all I did was set up an account there, so I haven't had to deal with the complexity myself.
With ChatGPT Sites... the one app I've hosted there (www.CardValuator.com) seems to load pretty fast, but it's also pretty much a compact one-page sorta site :).
Perhaps for my next vibe-coding project, I'll check out here.now! I think it's cool that -- unlike Substack -- they support BYODomain without charge :).
I made a seamless pattern generator—seamlessify (https://seamlessify.com/). It eliminates image edge seams to create infinitely tileable patterns (also supports AI generation from prompts).
This is a great idea - design vocab is exactly where I get stuck prompting agents. One related gap I keep hitting: once the designs are generated, I often need to merge two images (a layout over a background, before/after pairs). I've been using https://aiimagecombiner.app for quick composites instead of opening a full editor.
Ben, re: “I kicked this off in Pi using Fable 5.1 - I just wanted to brainstorm and play with a few ideas first, when I wanted to build it for real I used Factory’s Droid.”
Curious, why Pi to Droid?
its super context light and gets out of the way - so to have the clearest ‘mind’ i use pi. with other coding agents their system prompt gears them up for coding tasks.
its like sometimes if i need a less agent-y agent i’ll just ask plain http://claude.ai or http://chatgpt.com sometimes even with icognito to make sure no context is misleading it
Also, thanks for the sharing. Appreciate the behind the scenes, thought process, decision making, explorations, etc.