{
  "evidence": {
    "runId": "9b227585-e2f9-4d31-adff-e0113979baa1",
    "evaluatedAt": "2026-09-27T07:56:29.897Z",
    "packageVersion": "0.7.0",
    "recipeFingerprint": "e3afba589f57bf106ce32a86e6d6dd3eb7d42c3c7f43a4e659330e8368a2cad6",
    "datasetFingerprint": "3dd4a151b326da19ec7178454fe1e6bb72955792934a6a0c26c7644ffb8da4df",
    "split": "held-out",
    "mode": "live",
    "policy": {
      "minConfidence": 0.8,
      "selectedOn": {
        "runId": "2928ab8e-ec62-4300-bf44-b690425a5340",
        "datasetFingerprint": "df3494b970c0e117b7f35b64170beb57509e9d6d2a574617afa162c1d8bb5ed8",
        "recipeFingerprint": "e3afba589f57bf106ce32a86e6d6dd3eb7d42c3c7f43a4e659330e8368a2cad6",
        "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": "callback-responsibility",
  "model": "jev-1.13.0",
  "models": [
    "jev-1.13.0"
  ],
  "cases": 40,
  "completed": 40,
  "failed": 0,
  "correct": 39,
  "wrong": 1,
  "ready": 24,
  "review": 16,
  "accuracy": 0.975,
  "completedAccuracy": 0.975,
  "accuracyInterval95": [
    0.8712,
    0.9956
  ],
  "readyAccuracy": 1,
  "readyAccuracyInterval95": [
    0.862,
    1
  ],
  "reviewRate": 0.4,
  "contestedAccuracy": 1,
  "adversarialAccuracy": null,
  "calibration": [
    {
      "confidence": "0-0.5",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.5-0.6",
      "cases": 1,
      "accuracy": 0
    },
    {
      "confidence": "0.6-0.7",
      "cases": 2,
      "accuracy": 1
    },
    {
      "confidence": "0.7-0.8",
      "cases": 2,
      "accuracy": 1
    },
    {
      "confidence": "0.8-0.9",
      "cases": 5,
      "accuracy": 1
    },
    {
      "confidence": "0.9-1",
      "cases": 30,
      "accuracy": 1
    }
  ],
  "thresholdEvaluation": "frozen-policy",
  "thresholds": [],
  "suggestedMinConfidence": null,
  "latencyMs": {
    "p50": 112.84,
    "p95": 164.19,
    "totalCaseTime": 4717.47
  },
  "usage": {
    "requests": 40,
    "input_tokens": 27344,
    "output_tokens": 2133,
    "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": 0.00114845
  },
  "confusion": {
    "{\"verdict\":\"business\"}": {
      "total": 10,
      "correct": 10,
      "misses": {}
    },
    "{\"verdict\":\"customer\"}": {
      "total": 10,
      "correct": 9,
      "misses": {
        "{\"verdict\":\"business\"}": 1
      }
    },
    "{\"verdict\":\"either\"}": {
      "total": 3,
      "correct": 3,
      "misses": {}
    },
    "{\"verdict\":\"none\"}": {
      "total": 4,
      "correct": 4,
      "misses": {}
    },
    "{\"verdict\":\"unclear\"}": {
      "total": 13,
      "correct": 13,
      "misses": {}
    }
  },
  "failures": [
    {
      "id": "featured-16-business",
      "expected": {
        "verdict": "customer"
      },
      "actual": {
        "verdict": "business"
      },
      "confidence": 0.56
    }
  ],
  "acceptance": {
    "minConfidence": 0.8,
    "minimumHeldOutReadyCases": 20,
    "minimumReadyAccuracy": 0.95,
    "maximumProviderFailures": 0,
    "met": true,
    "label": "Measured on these synthetic cases"
  }
}