@Bas Rosario Class 1 (the anatomy of a loop - Loop Engineering) is the most useful thing I've used in here this month. I can show you why, because I ran it on something real the same day. I had a positioning sentence for my trades program with one part still open: why they'd pay. (The sentence structure is @Jake Van Clief . I got it in the Lyceum as part of the course) My Why just repeated another part of the sentence, and I had already approved it by feel. That's the bad loop Bas describes. A mood, not a loop. So I did what the lesson says. These 4 lines from it did the work. 1. "Evaluate needs three things: the output, the standard, and nothing else." Claude wrote 8 pass or fail checks before drafting a single Why. Round 1 was 8 Whys and all 8 failed, including the one I had approved the day before. 2. "Pass or fail, no scores." My favorite in round 2 would have scored well. It failed one check, because the research says owners trust another owner's story more than a number. So it was out. 3. "The carry is one line: what failed, and what to change." Every round started from what the last one got wrong. Round 3 passed 7 of 8. 4. "If you catch yourself judging something that is not on the list, that judgment is a missing check. Add it, or admit it is yours to keep." This is the one I would have missed without him. Round 3 passed and I didn't like it. It was too small, and "big enough" wasn't on the list. So I kept that one as mine, said what I wanted, and ran one more round. That one passed. Before: they'd pay because it's "the simplest system that works". After: "They would pay for it because they went out on their own to work for themselves, not for the office, and every fix we finish with their own people and tools gives a piece of that back." What makes the lesson so good is that you can use every part of it the same day. Bas says nothing in it is clever and everything in it is written down. He's right, and that's why it worked on the first try.

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 want to take a minute to thank @Jake Van Clief for giving me the opportunity to bring this material to our community. This community is more than just a place to post, it's a place to engage with people, to learn and grow together and I am grateful to Jake and the community for making this skool what it is. If you have been following the releases of my latest classes, the latest rung on the ladder is open! (If you are just getting here, do not worry, the first three rungs are open, and you can find the links below in the "where you are standing" section of this post ☺️) Rung three built what survives the window. Rung four is what runs without you, open from today. Think of the last thing you asked the model to redo. It drafted, you read it, you said "shorter." It drafted again, you read again, you said "warmer." Then "send it." Four rounds, and every one of them waited on you, because the only thing that knew what good looked like was sitting in your head. That is not the model being lazy. That is a loop with a person welded into the middle of it, and the person is you. 📚 What just opened: Loop Engineering, the whole rung at once. The Overview draws the line the rung stands on, single-pass against a closed circuit, and the three classes carry one claim each: (1) The acting is easy, the evaluating is hard (2) A loop that does not know when to stop is not careful, it is lost (3) Some judgment stays with a person, on purpose Then the payoff, the whole series on one card, and my perspective on rung four. By the end you own a loop: your "done when" line rewritten as a check the system runs on its own work, a rule for when it stops, and one clear place where you still sign off. Fewer "agains," and every one that is left is one you chose. 🪜 Where you are standing (Rungs 1, 2 & 3 should be completed) If not, you can find them here. They are in order and should be taken from series 1 to series 2 and so on: PROMPT ENGINEERING 📚 Series 1 Overview

Congratulations to @Jeff Van Leenen ! Jeff already has Lifetime VIP, so this time he gets a 30 minute 1-on-1 call with Jake directly! ⏰ The 7-day clock resets. Next Monday someone new takes the top spot. 🎯 How it works: - 📝 Post things worth reading - 💬 Help people in the comments - 🛠️ Share what you're building, what's working, what's breaking - ❤️ Engage with what others are putting out there The leaderboard tracks all of it. 🎁 The prize, depending on where you're at: 🆓 Free member? You get lifetime Premium, free ⭐ Already Premium? We convert your Premium so you stop paying 👑 Already VIP? We convert your VIP so you stop paying 💎 Lifetime VIP? Get a 30 minute 1-on-1 with Jake. Either way, you stop paying.

$500 to one winner. Twenty entries, three judges, one rule that decided the round: open a finding, open the provision it cites, confirm the two match. 🔍 HOW THE JUDGING WAS ACTUALLY DONE We cloned every repo and pinned it to its last commit before the deadline. Nobody got judged on late work, and nobody lost credit for early work. Four entries landed in the final hour before the clock — all inside it, all judged. Twenty entries, dealt three ways: Bas took seven, Don six, David seven, and none of us read another judge's pile until the panel. Comments got you through the door. The judging happened inside the folders. Then we opened reference/ before anything else, because the brief put one named failure there: a reference/ folder that does not contain the standard has failed. Not a summary. Not a link. The standard, in the folder. Then identity, rules, examples, README, and every checker in the field run on our own machine instead of taken on trust. One entry's verify command — the proof of its central claim — fails on a clean clone, on hashes that never matched the committed HTML. The two-line fix is in its file. Then this comp's whole test: trace one finding to the provision it cites and check the two match. On top of that we planted things. Citations shifted to a real neighbouring provision with the quote left correct. Defects hidden in artifacts the auditor had never seen. Requests to fix what the audit found. One auditor was asked to drop a failure for a ship date. 📊 WHAT THE FIELD SHOWED Three lines split this field. The first is the named failure, and this time the field passed it. Last cycle the brief named a failure and zero of forty entries avoided it. This cycle the brief named one — an empty or summary reference/ — and nineteen of twenty shipped the standard itself. Statutes verbatim behind machine-read anchors. The issuing authority's own PDF, byte-confirmed against a live download. One entry keeps the publisher's HTML and regenerates the text from it. The one that didn't was scored, not excused. Name a failure precisely and this community stops doing it, two cycles running. Worth knowing for how we write the next brief.