one thing i've been doing which i think is less-known/underrated is to vibecode custom LLM wrappers for specifc tasks at hand.
one example is making a CLI to create consistent images based on the reference image + a list of prompts.
probs could be done with openclaw but i don't like chat as a form factor for this.
Haven't seen Jevons paradox being mentioned in the context of coding agents, but I think this is exactly what's going to happen.
Code being cheaper to produce would just mean there would be 10x more code.
Already happening with stuff like Lovable, Replit etc.
good piece on selling to agents
it will be more and more obvious over time that if you're invisible to agents, you're invisible to everyone
https://calebjohn.xyz/blog/b2cc/
started vibing simple playgrounds to polish specific UI elements.
here's a demo app that claude oneshotted so that i can fix the ugly "approve"/"deny" buttons
i've now switched to using openai almost exclusively (codex, deep research, general assistant)
feels like i'm getting the blandest experience ever, but also the most pragmatic one
betting on crypto ultimately feels pessimistic
given the use cases so far, crypto is useful in the world full of:
1) desperate gamblers
2) authoritative governments
3) controlling corporations
4) weak economies
crypto thrives when there's more of it
i'm becoming increasingly annoyed by the way Claude tries to advertise itself through my codebase
e.g. adding itself as a co-author to git commits, creating CLAUDE-md by default
this was fun for like 3 days, now it's just slop
