Something interesting about the age of AI, is the realization that most ideas are crap.
Because now you can build basically anything. And people do.
But the vast majority of it ppl either do not want, adds no value, or cannot be marketed.
@levelsio@_mattwelter@pedrotech_ The red queen effect is now in full swing.
For those that don’t know from Alice in wonderland the Red Queen tells Alice, “My dear, here we must run as fast as we can, just to stay in place”
Something I appreciate about Open AI & their usage limits: they dont kill a running autonomous codex session mid way.
Anthropic kills everything mid session. Even a running workflow w/ millions of fable tokens in progress gets burned, need to start it over later. It's terrible.
@grkn I think this is the problem, because they kill all the sub agents even if you want to continue it with credits you need to start all those workflows all over.
It costs a fortune and only they benefit. Very consumer unfriendly.
@johnennis If they wanted to do something to promote the product like
“Produced using claude” or something like that, maybe, but co-author?
Get outta here with that nonsense
@petergyang Personally, I think this is just the endless Server vs Local cycles in software.
Like how we used to run terminals connected to large servers in the 80s. Or the SaaS outrage a few years back, firms going SaaS just because etc.
Imo both local & hosted models are important
I mean just think about how crazy this all is.
You stand before what is made out of mostly rocks (silicon is extracted out of quartz rocks) and you say "work on this for me" and it does
And if you use the right words, it'll get really close to giving you what you want
Made a little device that manages my claude code’s & codex’s on my server…
Simple esp32 s3 chip by waveshare. The orchestrator agent works with voice, controls my device remotely with tools, and controls the server.
It’s a cool time to make stuff.
@xtrasmal Yup, it’s wifi (but that chip has bluetooth too)
VoIP on the server, experimented with open TTS/SST/LLM on server it connects to, but also all eats resources/compute I want to use to do work etc.
Outside it uses phone as wifi hotspot, so same latency
@DenLoginoff Voice agent is gemini live sdk, but could probably use any of em, eleven labs etc.
For sure I had to write that UI, but claude helped me a ton. Also wrote the games and all the entire firmware on the device
@lucasarruda Yea, just wanted to create all from scratch and have some fun.
I started with my own screen and just the chip, but didn’t feel very well made. But this is. Even wrote some games with claude in C for the esp32 :)