@theo you lowkey pushed me to do something I wanted to do already.
basically every prompt is a "commit"
you can use it to share states across agents
or just recover a good session
Forgive my audacity here, but I kind of ALWAYS wanted this since day 1 of vibe-coding.
The prototype was really damn easy and I'm already benefitting from this immensely.
I finally can close terminals without a care because I can just ask my agent to look for the hist if i want
Ok, hear me out...
Every.
Prompt.
Is a
Commit.
Every prompt and every answer is accessible by any and every agent, any and everywhere.
All with refs to the file changes.
Let me cook
I'm going to use my AI psychosis to fix clouds for agents.
Someone else needs to use their psychosis to fix source control. I would do it myself but I'm already too deep on the cloud thing.
GitHub is dying and git is not the right primitive. Will dump some thoughts here.
@theo@JamesCanaryII I lowkey feel honored that I made it to this.
Your live (& video) about growing online presence definitely had an impact.
I'll keep trying to post useful and interesting stuff ๐ซก
@duedillengence@CryptoMikli I've been doing this for 4 years, but if I excluded my gains from before because they're WAY higher than 16%
Because I did my correct research and picked companies.
He said we can't outperform by picking up our stocks, I did, that's it
@CryptoMikli wanna beat it too? Look at Take-Two's stocks.
GTA VI will sell more than all the 3 Avatar movies together on the first month alone.
Even if it only sells 1/3 of what I expect, that game is a goldmine.
I trained my agent with the knowledge of only investing in good companies and identifying seasonal or strange dips.
It keeps printing me money, but it's not eager, it knows sometimes we just need to wait.
I'm down on only 1 stock (JP Morgan), which WILL go up.
I have beat it 4 years in a row, still 100% win rate.
idk if he's generalizing, but I guess I agree with the principle: stick to good stocks.
Even if you're down 20%, you will recover. Don't sell at lost and you'll still be over 15%/year which beats the S&P
Kevin OโLeary says almost nobody can beat the S&P 500 by picking individual stocks
โIf you think youโre so good that you can pick stocks and beat the index, give it a try and learn the hard way. Virtually nobody beats the indexโ
โWhat you should do is make it easy for yourself. Maybe put $1,000 into one account where you pick stocks, and another $1,000 into SPY, the ETF that tracks the entire S&P 500โ
โAnd then youโll find out the hard way that youโre not that good. 90% of the time, you canโt beat it, so you might as well join it and stop trying to pick stocksโ
@shadcn@nishimiya agreed, but I actually don't minimize because the behavior is different, it throws the window into a side bucket that I can't see anymore when I try to see all windows with 3 fingers up on the pad
@theo Sadly, yeah. It's just easier to keep track and separate projects, and sometimes I just want a quick "claude, do this" instead of having a whole workspace tool
@unclebobmartin@kubasienki Well, I guess "discipline" in software engineering is more important now than ever was, devs just didn't have time for all that
So I'll give you that