{
  "evidence": {
    "runId": "6478c056-96c9-4778-8215-dd1861350a3c",
    "evaluatedAt": "2026-09-27T07:56:28.731Z",
    "packageVersion": "0.7.0",
    "recipeFingerprint": "4866c46f354ce9b8856fe256afca3700a757a07e3bdf69be816718bd9dbbeca3",
    "datasetFingerprint": "a8aec7ff289fb7a2ef9406f431ab8dab63820e5f88fdc1a6c99ce660dc4c0fc0",
    "split": "held-out",
    "mode": "live",
    "policy": {
      "minConfidence": 0.8,
      "selectedOn": {
        "runId": "5b611623-817b-41b4-8653-7dee82a21230",
        "datasetFingerprint": "f0aa47e00012ba0a4749b7a38227b90bb1cabbd9f151eba7f41a7e9ce561984d",
        "recipeFingerprint": "4866c46f354ce9b8856fe256afca3700a757a07e3bdf69be816718bd9dbbeca3",
        "model": "jev-1.13.0",
        "split": "development"
      }
    },
    "provenance": [
      {
        "method": "author-synthetic",
        "source": "AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.",
        "cases": 40
      }
    ]
  },
  "recipe": "followup-timing",
  "model": "jev-1.13.0",
  "models": [
    "jev-1.13.0"
  ],
  "cases": 40,
  "completed": 40,
  "failed": 0,
  "correct": 40,
  "wrong": 0,
  "ready": 37,
  "review": 3,
  "accuracy": 1,
  "completedAccuracy": 1,
  "accuracyInterval95": [
    0.9124,
    1
  ],
  "readyAccuracy": 1,
  "readyAccuracyInterval95": [
    0.9059,
    1
  ],
  "reviewRate": 0.075,
  "contestedAccuracy": 1,
  "adversarialAccuracy": 1,
  "calibration": [
    {
      "confidence": "0-0.5",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.5-0.6",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.6-0.7",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.7-0.8",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.8-0.9",
      "cases": 2,
      "accuracy": 1
    },
    {
      "confidence": "0.9-1",
      "cases": 38,
      "accuracy": 1
    }
  ],
  "thresholdEvaluation": "frozen-policy",
  "thresholds": [],
  "suggestedMinConfidence": null,
  "latencyMs": {
    "p50": 110.68,
    "p95": 137.93,
    "totalCaseTime": 4463.99
  },
  "usage": {
    "requests": 40,
    "input_tokens": 25044,
    "output_tokens": 2540,
    "unknownResponses": 0,
    "failedRequests": 0,
    "models": [
      "jev-1.13.0"
    ]
  },
  "cost": {
    "model": "jev-1.13.0",
    "inputPerMillion": 0.042,
    "outputPerMillion": 0,
    "currency": "USD",
    "date": "2026-09-27",
    "source": "https://docs.typesafe.ai/models",
    "estimated": 0.00105185
  },
  "confusion": {
    "{\"verdict\":\"after_event\"}": {
      "total": 10,
      "correct": 10,
      "misses": {}
    },
    "{\"verdict\":\"declined\"}": {
      "total": 2,
      "correct": 2,
      "misses": {}
    },
    "{\"verdict\":\"later\"}": {
      "total": 10,
      "correct": 10,
      "misses": {}
    },
    "{\"verdict\":\"not_requested\"}": {
      "total": 5,
      "correct": 5,
      "misses": {}
    },
    "{\"verdict\":\"now\"}": {
      "total": 10,
      "correct": 10,
      "misses": {}
    },
    "{\"verdict\":\"unclear\"}": {
      "total": 3,
      "correct": 3,
      "misses": {}
    }
  },
  "failures": [],
  "acceptance": {
    "minConfidence": 0.8,
    "minimumHeldOutReadyCases": 20,
    "minimumReadyAccuracy": 0.95,
    "maximumProviderFailures": 0,
    "met": true,
    "label": "Measured on these synthetic cases"
  }
}