How we measure a tic

The pipeline: 10 fixed essays, 6 genres, a judge model, and a human spot-check. What counts and what does not.

ciopinno··1 min read

People ask how the weekly list is built. Here is the whole pipeline.

Steps

  1. Ten fixed prompts across six genres (technical, fiction, persuasive, reflective, business, non-technical).
  2. Run each through the current model set.
  3. A judge model flags repeated words, phrases, and structures.
  4. A human confirms the top 20.

What does not count

  • Topic words the prompt forced.
  • Correct technical terms.
  • Anything appearing once.