{
  "registry_version": "1.0.0",
  "data": {
    "title": "Take-home assignment exceeding reasonable stated scope",
    "status": "active",
    "evidence_ids": [
      "evidence.employment_interview_reliability_new_meta_analytic_estimates_by_structure_and_format"
    ],
    "content": "# Take-home assignment exceeding reasonable stated scope\n\nThe take-home assignment asks for production-grade features that take 20+ hours to build, against a stated budget of 2–4 hours.\n\n### Diagnostic Non-Inferences\n- Does not prove the company is harvesting free intellectual property for production use.",
    "aliases": [
      "A-006"
    ],
    "id": "obs.take_home_assignment_exceeding_reasonable_stated_scope",
    "type": "observation",
    "summary": "The take-home assignment asks for production-grade features that take 20+ hours to build, against a stated budget of 2–4 hours.",
    "stages": [
      "technical"
    ],
    "evidence_level": "supported",
    "probes": [
      {
        "id": "PROBE-A-006-1",
        "action": "Write the boundaries and trade-offs into the submission's README, naming the assumptions made to fit the agreed timebox.",
        "expected_signal": "Tests whether evaluators value scope management and communication over unbounded free work.",
        "cost": "low",
        "outcomes": [
          {
            "id": "readme-cited",
            "label": "Written feedback, or the next conversation, refers to the timebox note and the cuts you named, and treats them as part of what was submitted.",
            "excludes": [],
            "because": ""
          },
          {
            "id": "cuts-read-as-gaps",
            "label": "The reply lists the stubbed items as missing work, with no reference to the note that names them as deliberate.",
            "excludes": [],
            "because": ""
          },
          {
            "id": "template-rejection",
            "label": "A rejection in standard wording arrives with nothing in it specific to the submission.",
            "excludes": [],
            "because": ""
          },
          {
            "id": "no-reply",
            "label": "No reply arrives at all; the brief and the submission date are the only dated records you keep.",
            "excludes": [],
            "because": ""
          }
        ]
      }
    ],
    "specimens": [
      {
        "kind": "note",
        "label": "The assignment brief, as received",
        "subject": "Take-home exercise — Senior Backend Engineer",
        "context": "stated budget: 2–4 hours",
        "lines": [
          {
            "text": "Build a service that ingests order events, deduplicates them, and exposes a query API over the resulting state.",
            "tell": false
          },
          {
            "text": "Requirements: persistence with migrations, idempotent consumers, retry with backoff, structured logging, metrics endpoint, containerised local run, integration tests, and a short design document covering scaling and failure modes.",
            "tell": true
          },
          {
            "text": "Please submit within one week. We estimate 2–4 hours.",
            "tell": false
          }
        ],
        "reading": "Count the deliverables against the stated budget. The mismatch is measurable, which makes it a fact rather than a feeling."
      },
      {
        "kind": "chat",
        "label": "Asking about scope",
        "lines": [
          {
            "speaker": "Candidate",
            "at": "day 1",
            "text": "Happy to do this. To fit the 2–4 hours, which parts should I treat as the core and which can I stub with a note in the README?",
            "tell": false
          },
          {
            "speaker": "Recruiter",
            "at": "day 2",
            "text": "Ideally all of it — we use it to see how you work end to end.",
            "tell": true
          }
        ],
        "reading": "A direct question does not reconcile the stated budget with the expected deliverable. That is the observation; the reason for it is not visible from here."
      }
    ],
    "perspectives": [
      {
        "actor": "actor.candidate",
        "sees": "A brief with eight requirements listed against a stated budget of two to four hours, and an answer to the scoping question that leaves the list unchanged.",
        "reads": "The stated time and the stated deliverables are two different sizes, and which of the two the submission will be scored against is not written anywhere the candidate can reach.",
        "does": "Decides how many unpaid hours to commit, and submits with the timebox, the deliberate cuts and the assumptions written into the README, keeping the brief and the reply on record."
      },
      {
        "actor": "actor.recruiter",
        "sees": "A candidate asking which parts are core, and a brief that is the only document the recruiter holds about this exercise.",
        "reads": "A question about scope is a question about the rubric, and the rubric sits with the panel; the stage is measured by how quickly it clears.",
        "does": "Answers from the brief, since that is what the recruiter has, and returns the thread to a submission date."
      },
      {
        "actor": "actor.hiring_manager",
        "sees": "Finished submissions. How long any of them took is not a field in what reaches the panel.",
        "reads": "Differences in completeness read as differences in level, because completeness is the variable the submissions expose.",
        "does": "Compares the batch against itself rather than against the stated budget, and sets the bar from the fullest submission received."
      }
    ],
    "non_inferences": [
      "Does not prove the company is harvesting free intellectual property for production use."
    ]
  }
}