Been heads down on something the past month plus. Not a lesson, nor a Tea round but more like the layer under all of it. Kept seeing the same thing in comments and onboarding calls: People not knowing where to even start or if something’s already been covered somewhere in here. We're working on fixing that. Not ready to put it in your hands yet.

I built an agent that reads the AI landscape every day and hands me back a short brief. The part that makes it mine rather than a newsletter is the rubric, because it scores each item on how useful it is to me, not on how big the news is. What it reads - model releases, regulation changes, what the CEOs are saying, plus the newsletters and accounts I picked by hand. What it does with it - scores every item against a rubric built from my own projects, repos and goals. What I get - a short brief with the useful things at the top and most of it cut. The sources are not a general feed. I chose every one of them, so nothing arrives that I did not put in the list. That alone removes most of the noise before any scoring happens. The rubric is the interesting half. It knows what I am actually building, so a huge release that has nothing to do with my work drops down the page, and a small change that lands on a project I am in the middle of goes to the top. Two people running the same sources would get two completely different briefs. And when something reaches me that should not have, I tell it that and why. The reason goes into the rubric, so the next brief is scored slightly differently. Happy to go into any part of it if it is useful to anyone. [Attached is a sample from today's news brief, just for an example of the kind of stuff that it sends]

Just a simple poll interested in what people think and to spark some discussions....

I have finally realized I have two styles of ICM that fall outside of the five flavors of ICM in the classroom. There have been times that I have an ICM that is a factory of results that I use multiple times. Then there are times I use an ICM for a new skill/research/brainstorming. I get VERY frustrated when Claude moves ahead on a brainstorming ICM without me. So I have written specific instructions that I put into brainstorming ICM's to be interactive with me at each step, and that I will ask many questions to determine the direction. Two of my factory ICM's - Creating brand one pagers, and putting sample data into a salesforce sandbox. One of my Brainstorming ICM's - Building out a sample dashboard for a mentoring program in real estate. Steal this if it helps anyone.

I got Claude to fess up on a few things. I apologize if I am not sharing MY ALL CAPS PROMPTS that lead Claude into this reveal. It wasn't really the caps - it was more the curse words. Thought is was interesting to have Clause speak directly to its 'bias to run'