Someone made a comment in a post last week about something their friend said that keeping up with AI is for people who are unemployed. I will modify that with its for people who are unemployed or its their full time job. Just trying to keep up with the news and develop stuff is a full time job in and of itself. LOL I don't got the bandwidth to keep up with everything! I need to win the lotto or something so I can devote the time I want to do it.

Has anybody used both? Which provided better output, and which model are you using for design? Thank you 🙏🏼

If you've been in this community for a minute, you've heard the term thrown around. ICM, Interpretable Context Methodology. Maybe you've seen it in Jake's lessons. Maybe you saw someone's folder structure screenshot and thought "that looks organized, but I have no idea what I'm looking at." 😅 I'm going to break it down the way I wish someone had broken it down for me. What ICM actually is (in plain English) 📂 ICM is a way of organizing your AI work, so the AI only sees what it needs to see, when it needs to see it. That's it. Instead of dumping everything into one massive prompt or letting a framework manage your context behind the scenes, you use your filesystem, folders, markdown files, plain text, as the architecture itself. Each folder is a stage. Each stage has one job. A CONTEXT.md file at the top tells the agent what this stage is, what inputs it expects, and what output it should produce. The agent walks into the room, reads the brief on the wall, does its job, and leaves. The next stage picks up the output. No frameworks. No LangChain. No AutoGen. Just folders and files. 🗂️ If you want to understand the philosophy behind why this works, where all of this leads, Jake lays the foundation here: 0.1: Where All Of This Leads - The Foundation Why it matters 🎯 Most people hitting a wall with AI aren't hitting a model limitation. They're hitting a context problem. The AI is trying to hold too much in its head at once. It forgets things. It contradicts itself. It hallucinates. It gets "lazy." 😴 That's not the AI being bad. That's you giving it a 47-page brief and asking it to stay sharp on page 43. ICM fixes that by isolating context. Each stage only loads what's relevant. The AI stays focused because you've structurally made it impossible for it to get distracted. The 60/30/10 rule (the lens behind it all) 🔍 This is the framework under the framework. Jake breaks every system into three layers:

Hi friends. Last week was my first submission into the weekly comp. Although I had no illusions about winning (or even placing), I was excited to build and learn. I learned a lot about the build process, but would love to learn about what I could have done better. Since there's no feedback from the judges, I was wondering if anyone would be kind enough to look at my work, and give some constructive criticism. Your wisdom would be much appreciated! here's my submission: SHIPYARD is for the person who needs to learn AI but has no idea where to start. They've tinkered, hit walls, and stalled trying to figure it out alone. That was me until a few weeks ago. Then I stumbled upon the Clief Notes community. The knowledge here has unlocked something for me. But reading lessons and then trying to build on your own can still feel like fumbling in the dark. I'm sure a lot of you, like me, would love to have Jake sitting next to you — coaching you through every step. Shipyard is my v0 attempt at that. This tool grew out of the lessons taught in this very community — the same folder-structure and first-principles thinking that makes AI finally click. Shipyard is that coach. It takes someone from "I don't know how" - to proof, in their own hands, that they can build something real. Drop the folder into a Claude or Codex project and it becomes your coach. You pick one of four practice builds: File Organizer, Brain-Dump Planner, Notes, or Decision Helper. It walks you through making the tool as real files on your own computer, one step at a time. It teaches before it asks, pushes back when you're vague, and makes you write every line yourself. In about 20–30 minutes you walk out with a working AI tool that you built yourself, and actually understand. Landing Page: https://julianargus01.github.io/shipyard Repo: https://github.com/julianargus01/shipyard

Looks like along with the release of Opus 4.8, some changes were made to Claude Chat. Under the models, this morning for Sonnet, it was just Adaptive Thinking slider. Now it has an effort selector Low, Medium, High and Max, along with the adaptive thinking slider. It shows up on Opus also. Haiku stayed the same.