Finally giving my view on JEV in my classic history and fundamentals fashion but... This one is a little different. It's my first time making a fully animated, pixel art video, and I'd honestly love to know what you think of the style. Should I make more like this? I made an ICM that fully automates most of the process i was able to get this video built in about three or four hours. Talk about a fun animation style right? Might redo a bunch of the videos in the classroom with this.

I just watched the breakdown from @iscalelabs, and the biggest thing that stood out to me was the building floor plan metaphor. It clicked because it clearly showed how simple pathing stops AI from getting lost, without buying expensive software or burning through context tokens. My main takeaway was how a basic folder structure with an agent.md file gives you full control and consistent results. The experiment at the end helped me understand that when an agent is given a clean map, it can fetch exact info across different apps and models without hallucinating. Going forward, I'm going to build my own folder-based router system to act as a Second Brain and keep context locked in for all my projects.

I'm currently using AI to streamline web development, automate workflow processes, and deconstruct marketing strategies for my holistic health business. Right now, I'm at the building structured systems stage. One thing that caught my attention from this lesson is modular context management, because token efficiency and clear execution are crucial for me. I need to squeeze maximum utility out of AI without wasting time repeating instructions or burning tokens on unnecessary overhead. I'm excited to learn how to master Context Engineering to build lean, repeatable pipelines and see how structured approaches to AI can help me cut through the creative noise, stop reinventing the wheel, and actually get my holistic practice shipped and running.

CONTEXT ENGINEERING 📚 Series 2 Overview - THE LEGENDS · Clief Notes Picture your best instruction from rung one, word for word, handed to a new hire on their first morning, with no files, no sample of your work, no idea who the customer is, and nobody telling them what they may touch and what they may not. The instruction is perfect, the work comes back wrong anyway, and it is not the new hire and it is not the sentence. It is everything you did not hand over with it. That everything has a name, context, and as of today it has a whole rung. 📚 What just opened Context Engineering, the whole rung, seven posts. The Series 2 Overview sets the map, then three classes, one claim each: (1) A context window is a budget, not a bucket (2) Two examples beat a page of adjectives (3) The answer depends on what the model is allowed to touch Then the payoff, where all of it lands on one quick card you can save, then where I stand on rung two, and one short closing post that shows you where this fits in ICM. By the payoff you will have seven lines written down, one page you can paste into a fresh window on ANY day and get the good answer back (yes, the same prompt you walked in with, you will not change a word of it). 🪜 Where you are standing If you climbed rung one, you are ready, and bring your best prompt from that series with you, because the Overview puts it to work before class one even starts. If you have not climbed it yet, the bottom is still open and it is still the place to start. This rung stands on that one, the Start Here post is pinned and it is a short read, and I would rather you arrive a week late than arrive without it. PROMPT ENGINEERING 📚 Series 1 Overview - THE LEGENDS · Clief Notes 📆 One series a week, the same as I told you at the bottom. Rung three, Memory Engineering, lands next week, and then two more to the top.

Eight months ago, I started building what I thought orchestration required — N8N, Postgres, LibreChat, consulting and hosting fees. Tens of thousands of dollars. That was the right bet at the time. Nobody I saw was doing this with Claude yet. Three or four months ago, the game changed. And I didn't know it until I stumbled onto Jake's content on YouTube. Two weeks after watching "Stop Building AI Agents. Use This Folder System Instead", I have a working MCP. Client demo-ready. The whole system I was killing myself over? Markdown files and orchestration prompts. That's it. I want to give credit where it's due. Jake put something out that reoriented how I thought about AI. The investment wasn't wasted — it built the foundation. But Jake and this community pointed me toward what was actually possible now. If you're still building the complex version because you think you have to — it's worth a second look.