attempting to see how many new features i can write simulatenously while still being in the loop. it's like playing N games of chess at the same time.

nostr can uniquely solve a lot of ai data siloing problems, and it can make it privately shareable as well

Which brings me to my main complaint about opencode: you can’t swap between these modes easily like you can with claude code. Human-in-the-loop vs not. Bizarre design choice View quoted note →

I’ve always joked that a programmers job is to software hard enough to the point where we aren’t needed anymore. I don’t think this is a joke anymore

vibe coding is too binary of a term and is causing confusion. The way I’m starting to look at it is agentic coding with different levels of human-in-the-loop. Sometimes you want to be in the loop (pair/symbiotic programming) sometimes you don’t (vibe coding) vibe coding isn’t necessarily bad, its just bad for certain problems: ones where you should be in the loop. And yes i still think humans should still be in the loop. For now.

my vibe coded embedded renderer for notedeck is coming along. opus cooked on this one

another nice aspect of vibe coding I underappreciated: when I'm decompressing after work I can turn into "lazy pair programmer mode" and just play runescape and watch youtube while the agent is still doing stuff. i just have to pop in and review code chunks when it notifies me. i can work when I'm not working! View quoted note →

currently have ~3 claude code agents going through all my github issues and fixing simple things I don't want to fix

theory: the number of apps competing for peoples attention will only accelerate from here. you can’t launch mid things anymore. It will have to be exceptional and mind blowing.

can someone vibe code GTA6 i don’t wanna wait no more