Systems Thinking

You don’t need more AI you need a system otherwise you’re just moving faster in the wrong direction agree?

@jspeiser tools changed but who’s in control of the system now? you or the tool?

First week with zero OpenClaw. every routine, every automation runs on Perplexity Computer now. No command line. No babysitting. Time to service has cratered. I didn't switch for the hype. I switched because my time kept disappearing into a black hole. Now it doesn't.

@2sush using AI isn’t the flex either understanding it is 🙂

you don’t hav to love AI, but ignoring it isn’t a flex.

@sahill_og we’re not handing it over we’re trading it for speed at what cost?

People fear AI taking control. Reality: we’re handing it over.

@itsolelehmann 90% might happen but 90% control won’t who owns the decisions?

pretty sure in 6-12 months most people will do 90% of their work via claude interface

@dennysentinel @nalinrajput23 it’s not copy-paste anymore it’s blind approval same risk better UX 🙂 how often do you actually read what you ship?

@OnePersonSys @nalinrajput23 nobody is copy pasting anymore that was like a year ago.

Be real… does learning to code even matter anymore with AI everywhere?

@minzengnotes @ThoughtfulTechy hunger can break a system but without one you just burn out how many realize it?

@OnePersonSys @ThoughtfulTechy When there’s enough hunger, the force might change the system.

You only need two things to be successful in tech: 1. Curiosity 2. Hunger

@fing_me_gud Engagement is good, but it’s not enough. Build something real, and people will engage anyway.

If you follow me, I’ll follow you. If you reply to me, I’ll reply to you. If you engage with me, I’ll engage with you. Let's engage and support each other! 🚀🤝

quiet means B 🙂

be honest!

If AI writes your code but you can’t explain it are you still coding or just approving output?

@krzyzanowskim AI writes faster bugs ship faster 🙂

every other day I want to quit supporting claude code in the Commander - the SDK is either flacky, broken here and there with each update, or straight up lack support certain features

@lucas_montano building got easier understanding didn’t 🙂

claude and codex just made it easier to build native apps.

@MillionInt we’re not bad at measuring intelligence we’re measuring the wrong thing 🙂

Lots of inconvenience comes from the fact that we’re just very bad at measuring intelligence And we think we’re not

@burkov models arguing is the feature humans picking sides is the bug 🙂

GPT-5.4 > Opus 4.6 And Google still doesn't have anything even remotely competitive.

@zuess05 boring is the feature, not the bug 🙂

@OnePersonSys Ugh that’s so boring

You just got a god-tier idea for a app/saas. What is your literal next step? A) Open Claude Code and start working B) Build a waitlist landing page C) Cold DM 50 potential customers D) Tweet the idea to see if anyone cares Be honest.

While building my own runtime i realized most people don’t have one they just react what’s your intent?

@Bencera this is the real shift not AI replacing engineers but engineers orchestrating disagreement if your models never disagree, you’re probably not pushing hard enough

my AI coding workflow as a solo founder: - opus 4.6 for exploration + planning - codex 5.4 xhigh to stress-test the plan (catches gaps opus missed) - back to opus, which usually complains codex is overengineering lol - few rounds back and forth. codex implements, opus reviews. - ask both: "safe to ship? what's the worst thing that could happen?" opus and codex arguing over my codebase is my entire engineering team. will probably ship this workflow as a Polsia feature at some point.