Session Extractions
One session, end to end. An X post about fixing AI slop became a running quality gate, a course on why measurement is the moat when everyone has the same model, and a reasoning framework for building less. The course article and both teaching blocks.
Course
When everyone has the same model, generation is commoditized and measurement is the edge. A hands-on course for building an AI quality gate that compounds into a moat, with a no-code track through every step.
Read →Teaching Blocks
A working session where an X post about "fixing AI slop" became a real, running quality gate — one that blocked a bad change, caught a silent production dip, and turned every failure into a permanent test. The whole build, the verbatim scores, and why a measurement layer is what stands between your brand and the slop machine.
Read →A meta-analysis of the reasoning pattern under the build. At nearly every fork, the right call was to build less — delegate it, decompose it, defer it, or delete it. Six pre-build questions, the Subtraction Pass, that remove work before you write a line of code.
Read →