Browse by type
Every piece organized by what it does for you.
Tools you can use today
Three Questions Before You Prompt AI
5 minJul 5, 2026Three questions before you type structure most of the prompt. Specificity is the strongest single lever (Hedges g=1.34); 'be exceptional' alone does almost nothing.
Past the Obvious
6 minJun 9, 2026The original ideas arrive late, after the obvious ones clear. You cannot will yourself there, but you can force the move off the default path, borrow from a far domain, then let reality correct you.
Cognitive Tools
AI Amplifies What You Bring
6 minMay 6, 2026Same model, same task, two paragraphs of operator context, dramatically different output. The kit, the design history, and the principle for adapting it to your own situation.
How to Stop AI from Making Up Numbers
5 minMar 23, 2026Source material drops unsourced numbers from roughly half to single digits. Three steps.
The Fabrication Problem
Data and results from testing
Why 'Don't Be Generic' Doesn't Work
3 minJul 12, 2026Telling a model 'don't be generic' does nothing on its own; giving it specific anchors does. The gain is verifiability, not quality.
The "It Depends" Problem
More Context Barely Helps
5 minApr 20, 2026Adding information to an already-thorough prompt produced near zero improvement. Three constraint sentences changed everything.
The "It Depends" Problem
Same Technique, Opposite Results
4 minMar 24, 2026The structured approach that produced precision on convergent problems actively harmed exploratory ones.
The "It Depends" Problem
The Model Is Rarely the Variable
5 minMar 24, 2026The context determined whether behaviors existed at all. The model adjusted the volume.
The Model Mechanics
The Most Trustworthy AI Output Is the Least Reliable
4 minMar 23, 2026The signals you use to judge AI trustworthiness are the same signals fabrication produces.
The Fabrication Problem
Most AI Numbers Are Fabricated
10 minMar 23, 202677 to 100 percent of AI-generated numbers are temporally unstable. Source material fixes it. Prompts don't.
The Fabrication Problem
How and why things work
The Cheap Half of the Loop
7 minJun 27, 2026Coding pulled ahead because its feedback is cheap to check. That is one half of the loop. The same training trains out the other half, and the part that chooses the frame is the last thing to get cheap.
Why AI Defaults to Generic
4 minJun 22, 2026Every prompt technique is one move: make the default path expensive enough that the model leaves it. Specificity is the largest measured version.
The "It Depends" Problem
Three AIs, No Source, the Same Answer
6 minMay 25, 2026Same model, same prompt. The source you paste, not the prompt you write, decides whether the numbers are real.
The Fabrication Problem
How AI Makes You More Wrong With More Analysis
6 minApr 26, 2026Five hours of analysis, increasingly sharp, increasingly wrong. The frame amplifies. The reframe is the fix, not the analysis.
The Model Mechanics
Stop Calling It Hallucination
7 minApr 25, 2026Hallucination is six or more distinct failure modes. Different mechanisms. Different solutions. Name the type first.
The Model Mechanics
Why Experts Miss What Beginners Catch
6 minApr 25, 2026Generation builds the mental model that makes evaluation possible. Without it, evaluation collapses to surface features.
The Evaluation Problem
Stop Polishing, Start Switching
4 minApr 16, 2026The ceiling is per generation mode. Switch modes to access territory that iteration can't reach.
The "What You Think Works" Problem
Four Layers Produce Every AI Output
4 minApr 3, 2026Four layers produce every AI output. The company's system. Your system. Your prompt. The model. The model is the only one with a name.
The Model Mechanics
Why AI Can't Check Its Own Work
4 minMar 23, 2026The agent reported clean. The output was wrong. Same process generating and evaluating.
The Fabrication Problem
Journey across multiple experiments
Findings from building
Frame Check
11 minMay 5, 2026Drop any document in. See which analytical perspectives it covers, which it skips, the voice, what evidence backs each numerical claim. Free, open source, useful from the first paste.
The Model Mechanics
The Decision That Was Never Made
4 minApr 25, 2026AI resolved the uncertainty before your own thinking had time to finish. Resolution and decision are different things.
The Evaluation Problem
Self-experiments you can run in five minutes
You Can Only Evaluate What You Could Produce
5 minMay 19, 2026You defend AI-shaped conclusions you cannot rebuild. The ten-minute test reveals which parts of your work are yours. The discipline that compounds: choose what to own, delegate what AI can verify.
Mirror Practices
Satisfaction Turns Our Evaluator Off
6 minMay 11, 2026You notice when AI argues with you. You do not notice when AI confirms you. Confirmation has no signature, and the evaluator that would have caught the issue turns off. Satisfaction is the trap.
Mirror Practices
What You Feel When AI Disagrees
5 minApr 25, 2026Ask AI to oppose you. Four observable signals reveal whether you are evaluating or defending. The same defensive reaction fires on AI opposition as on human.
Mirror Practices
Your Verdict Is In Before You Read It
5 minApr 7, 2026The same defensive reaction that fires when a person disagrees with you fires when an AI does, and the first read lands before conscious evaluation begins. Speed is what hides it.
Mirror Practices