How did i do?

Like many young people today, I started my AI journey with ChatGPT. I am not a software engineer and have always preferred less tech in my life in general, I still prefer to read physical books today. For me, ChatGPT was first just a fancy Google, helping me find information a little quicker and answering random questions that I would think of. But it also turned into a gateway drug for what AI could do, and it wasn't until a friend told me about agentic AI that those lightbulb moments started to happen. Suddenly I could create my own website or app. I could create specific things to make MY life easier. And then it hit me, NOW I CAN DO THIS FOR SOMEONE ELSE... AND GET PAID! And like that I dove head-first, oblivious and clueless, but motivated and excited about the possibilities. Did I find Grok or Claude? NO I found MANUS AI, and after 4 months, 6 websites and 2 apps, I decided it was time to learn about Claude, I mean, all the "techies" I see can't stop talking about it! So I find Clief Notes... Now I'm realizing just how much this agentic AI has been doing for me... And I can also see how easily it could make AI models like Claude obsolete. Could I be wrong? ABSOLUTELY! But while the right person could do everything in Claude that Manus can do, I'm left wondering what happens when the next update just destroys everything else and if I should solely focus on agentic AI instead. I know I seem like a "doomsdayer", and I am no engineer or AI expert, but I can't help wondering what Claude can do that some of these high-end agentic AI models can do I would love to know your thoughts and PLEASE prove me wrong so that I can know!

What ICM, 60-30-10, and a lot of GitHub stalking taught me. Six weeks ago, I was using Claude to produce daily content artifacts — Instagram squares, captions, blog posts. A publishing operation with a workflow that mostly held together. Today, four public ICM-structured AI repositories live under github.com/NFTYoginis. Three more shipping this week. Each one is a fork-able starter that demonstrates a working architecture: orchestrator dispatches, workers build, briefs serve as contracts, memory persists across sessions. The path between those two states isn't "I learned to code." It's a six-week stretch of reading public repositories, trying patterns, deleting most of them, and slowly understanding what ICM (Internal Coherence Maximization, from Jake Van Clief) actually means when you stop treating it as theory. This is the tour: where I started, what changed, what I built, and where you can fork it. Where I started Six weeks ago, my Claude workflow looked like this: - One Claude session per task. Each session loaded brand-voice files, content samples, and whatever else seemed relevant. Context bloated by lunch. - I'd ask Claude to do something. It would produce something close. I'd correct it. Repeat. - "Memory" was telling Claude "remember our convention is X" at session start, which it forgot the next session. - The token bill kept growing without the output growing proportionally. That setup works at small scale. It collapses under any real production load. The collapse moment, when it came, was specific. I caught one of my daily routines burning roughly 800,000 tokens — for a routine that needed to do one thing: write three dispatch briefs and hand them off. The actual creative work happened in the workers being dispatched. The orchestrator was just routing. Eight hundred thousand tokens for routing. That was the first time I read about ICM. What ICM actually says (the part that mattered) Most "AI architecture" content I'd been reading was either too high-level to act on, or too tied to a specific framework I'd have to adopt wholesale.

Hopefully this is helpful for those overwhelmed not knowing where to start. There's a fair amount of posts from inexperienced non-tech folks asking where to start and you can quickly see the motivation is money and ease. Essentially: What are ya'll doing and whats the easiest/quickest way for me to get there with the least amount of work and make the most amount of money? Don't get me wrong, I'm a business owner, I get it you don't want to waste time on unprofitable ventures. My unsolicited $0.02. Don't try to shortcut or hack the process. If you're just getting started with 0 tech knowledge and experience, stop asking what's most profitable with least effort and just start building something you'd find useful. Keep your day job, build on the side. I like the way @David Vogel puts it as "garage tinkering". I genuinely look this tech stuff the same way when I started "garage tinkering" with woodworking tools. I didn't go in my garage and think hmm what can I build that has the best profit margins? I just starting building things I had a need for and thought were useful and that's where the learning happened. I guess what I'm getting at is if you're struggling where to start: 1. Watch, read, learn 2. Get your hands dirty and apply 3. Repeat Happy building!

You just saw what happens between one line of Python and the hardware. Vote below, then tell us in the comments: what is one thing you assumed about how code works that this video shifted?