So, starting completely from scratch... The first thing I did was actually not build anything. 😄 I looked around. I started with the Classroom, specifically Getting Started and the first lessons of The Foundation. If you are completely new here, go and do that. It’s time well spent. Then if you want come back here. When I approached Module 4 in The Foundation I had a few questions. Mind you, at the time it was not as nicely organised as it is now. It has improved. Thank you @Jake Van Clief Anyway, the questions appearing in my head went something like this: How do I pick the right folders? Where do these .md files come from? Do I write them? How would I know they have everything inside them that my project needs? What even is my project? Can I have multiple CLAUDE.md or AGENTS.mdfiles? Where is my real folder root? Do I have multiple projects in the same folder? Can I add more stuff later?... Sound familiar? 😄 In the end, it all boiled down to one question: How are people actually doing this? 🔎 So I stopped the Classroom for a while and the search bar became my best friend. The more I searched and the more I looked around, I slowly but surely figured something out: There seems to be more than one way of building an ICM. Some members loaded Jake’s & David's ICM material/research paper into their LLM of choice and built from there. Other members started with the icm-architect skill. Others did a combination of both. And yet another group used the icm-architect together with something called the Workspace Blueprint. The bottom line for me was: ❗Many roads lead to Rome. Pick one. ❗ While combing through the group with my best pal, the search bar, I also came across some really interesting videos and posts which helped clear things up for me. Some were easy to find. Others were hidden away somewhere in the dusty abyss of the Skool community posts. 😄 In hindsight, I think searching around for a while was actually very useful. It helped me connect the dots and, more importantly, stopped me from immediately jumping into building something without really knowing what I wanted.

If you opened this post, you might already have a very strong opinion about it. Let's be honest, it is practically every where (and I know I resort to it often). I was going to write a different post all together, when it occurred to me that it sure serves a section of humanity who has lost their voice, and who want to put content out. I wrote an opinion about it - https://thequietai.com/writing/say-whats-synthetic.html Though a poll is far more fun -

I was invited to sit on a panel this morning: "AI in Action: The Mindset Behind the Moves," at Auxiom's Community Appreciation Event, attended by 120 business leaders in Detroit Metro. Grateful to Founder and CEO Matt Loria for the invite. Auxiom is a local IT and cybersecurity company. Before the panel, Ashok Sivanand (Nous) ran the room through a workflow-redesign workshop. He opened with a stat from McKinsey: 80% of companies doing generative AI work saw no change in profitability. He paired it with another: only 1 in 5 companies had actually redesigned a workflow before pointing AI at it. Same 1 in 5, both times. He called it a coincidence. This is similar data I referenced in David's Corner, The Tool is 10 Percent. His worksheet was amazing though: name the constraint, name the workflow, name your customer, find the leak, pick one fix, run a 30-day experiment with a guardrail and a report-back date. His line on the handout: diagnose before you prescribe. All 120 execs walked out of that room with a real next step; that's rare for many speakers. On the panel, I got to talk about an idea I've been writing about here: AI does two different jobs. It's good at mining your own record for what's already true, and it's good at amplifying whatever you give it. Skip straight to amplifying, and the output sounds like everyone else's. Someone asked what my own transition looked like, going from a 12-person team to running alongside 20-plus specialists. Two years chasing tools before I found ICM. Two weeks after that, I converted all my unstructured data into structured markdown and built my routing map and orchestrator, Duke. After, a CIO on the panel asked me how I was able to orchestrate across 50k markdown files. I said, "Simple: a map, and dropped Jake Van Clief and ICM as the solution." The closing session was Aaron Salko and Kelly Johnson on human performance. A line I wrote down: "Technology creates capability; human performance determines the outcome". If you've got performance gaps today, AI doesn't fix them. It multiplies them across more people.

I think I have 30-40 hours left of Fable access. My usage has just ticked over to 56% used. I had enough of a build queue that I decided to upgrade my plan just for this month and the Fable-in-subscription window and honestly, it's already been worth it. How those tokens have been spent: - 5% used to complete a system recarve and hardening pass - 35% used to orchestrate a full substrate port (from NotePlan to Notion) *and* a full branch that is future compatible with that state ready to switch to on Friday assuming this week on Notion goes as planned (oh and a full school comms rewrite to go with my brand new gmail set up exclusively for my system to access). I didn't have to build a thing in Notion - Sonnet dispatches did the lot. - 5% used to build the work order stage of my construction crew - 11% used to build the orchestration, build oversight and dispatch stage of my construction crew (and a massive shout out to @Ari Evergreen - I forked Pushing Dispatch_ and that helped me keep this to 11% of Fable orchestrating, otherwise it would have been more) Every stage of the construction crew has been pushed through a live run of the step before it in the chain to produce the document that guided the build. For the work order stage, that meant a PRD from the first stage and a build spec produced by the second stage. For this orchestration, it meant that I had fully sliced work orders, mapped to my schema, clustered to work packets that made sense. 8 stages; that were able to run in an orchestration-style mode (on Fable, I would not have let Opus go like that to be honest) and I came back to high quality work, measured to my own rules and standards and only a few small tweaks that didn't require substantial rework before committing. Honestly, the fact that something of this scale could be overseen with only 11% usage feels like a real testament to the whole chain here. I mean... this is really something. That means I still have 44% of my Fable tokens to go and only 30-40 hours to spend it in... and now we see how much of it I can use before I lose it... Wish me luck.

This was the first community I joined, and my first real go at AI beyond just talking to a chat window. I found it pretty much by chance, lurked for a few months, and High Tea 6 was my first call. I didn't realise at the time how much that would shape things. The picture I got here was a healthy one. Build the thing, watch how it actually behaves, be honest about the bits that don't work. I've spoken to enough people since who came in through paid courses to know that isn't the norm. Nobody here sold me a shortcut, and I got pretty lucky landing here first rather than somewhere else. Something happened recently that made me take a step back. I met someone at an open networking event who runs corporate training and is writing a textbook alongside it. What she needs is her course turned into per-module walkthrough videos - her own slides on screen, voiceover over the top, one video per module. I built it as ICM, because that's just how I build now. Numbered stage folders, one agent reading the right files at the right moment, markdown carrying the context, and local scripts doing all the mechanical work that never needed a model in the first place. Her voice cloned from a module she'd already recorded, the deck walked and exported as frames, one script and one audio file per slide. The durations of those audio files become the cue sheet that stitches the whole thing together, so there's no separate timing step to get wrong. One module's through it end to end. 21 slides, 21 scripts, 21 audio files, one finished MP4. And there's around ten hours of modules to put through the same pipeline. As an 18-year-old that's a really good opportunity, and it's going to generate me some serious income. I wouldn't have built any of it without what I picked up in here first, so thank you to everyone who's answered a question of mine along the way. Nothing's finished yet. I've still got a lot to come and plenty I haven't cracked. It just felt worth stopping on rather than quietly getting on with the next thing.