ICM Changed How I Build. Here's What It Actually Is and How to Start.

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:

⚙️ 60% is deterministic infrastructure. Code, schemas, folder structure, contracts. Things that don't change and don't need AI.

🔀 30% is orchestration. Rules, templates, routing logic, stage contracts. The guardrails that keep everything moving in the right direction.

🧠 10% is AI judgment. The actual creative or analytical decisions only AI should make.

If your agent setup feels fragile, unpredictable, or expensive, check your ratio. You're probably letting AI do work that belongs in the 60% or 30%.

How to actually start 🚀

You don't need to overhaul everything. Start with one workflow you already do repeatedly.

Pick something like writing a blog post, onboarding a client, scoping a project, processing research.

Break it into stages.

What happens first? What comes next? What's the final output?

Make a folder for each stage. Number them. 01-intake/, 02-research/, 03-draft/, 04-review/.

In each folder, create a CONTEXT.md that answers three questions: What is this stage? What input does it expect? What output should it produce?

Run through it once manually with your AI. Drop the relevant files into each folder as you go. See what happens when the AI only has to think about one thing at a time. 💡

That's your first ICM workspace. It'll be messy. That's fine. You'll feel the difference immediately.

The full lesson library is here if you want to go deeper: CliffNotes Classroom. And if you want to see how Jake thinks about this stuff in short form, follow him on TikTok 📱

What I want to hear from you 👇

I'm working on a deeper breakdown of wins and losses, what actually improves when you implement this and where it gets tricky. If you've tried ICM or you're about to, drop your experience below. What clicked? What didn't? What are you stuck on?

The best way to learn this isn't watching, it's building. So let's build. 🤙

❤️‍🔥 Bas

We learn together, We grow together, We win together! 🏆

This content was helpful

All of the above!

6 votes

4 comments