{
  "ok": true,
  "data": {
    "entry": {
      "id": "eval-sandwich",
      "kind": "method",
      "kindLabel": "Method",
      "title": "Use the eval sandwich: before, during, after",
      "body": "One eval at the end is a smoke alarm, not a control system. Run a small eval before the change, a targeted eval during the change, and a regression eval after. The sandwich catches drift while the work is still cheap to fix.",
      "tag": "evals",
      "date": "2026 · 06 · 20",
      "publishedAt": "2026-06-20",
      "note": "The middle eval is where most prompt regressions get caught.",
      "insight": {
        "aha": "The most valuable eval is often the small one you run while the prompt is still being edited.",
        "tryThis": "Run a baseline set before the change, five targeted fixtures during the change, and a regression set after.",
        "watchFor": [
          "Only testing the behavior you intended to improve",
          "Skipping refusal and edge fixtures after style changes",
          "Treating one green benchmark as proof of production safety"
        ]
      },
      "extended": [
        "Before-change evals establish the baseline you are trying not to lose. During-change evals are small and sharp: the five examples most likely to break because of the edit. After-change evals make sure the fix did not damage neighboring behavior.",
        "This is especially useful for system prompt work because every instruction edit has a shadow. You improve one behavior and accidentally move refusal, tone, verbosity, or citation discipline.",
        "Treat the eval sandwich as a change-management ritual. It is less glamorous than a benchmark, but it is the thing that keeps production behavior from drifting by inches."
      ],
      "url": "https://tips.platphormnews.com/entries/eval-sandwich"
    }
  }
}