PII Removal

Detect and redact PII from documents using AI-powered analysis.

Enterprise plans. A request from another plan is refused with 402 naming the plan that has it — the same body as every plan-gated feature (documented under the status codes in the OCR reference).

POST /v1/pii/redact

Upload a document for PII detection and redaction.

Endpoint

POST https://api.deepread.tech/v1/pii/redact

Authentication

Include your API key in the X-API-Key header:

X-API-Key: YOUR_API_KEY

Request

Headers

Request Body (Multipart)

Example Request

curl -X POST https://api.deepread.tech/v1/pii/redact \
  -H "X-API-Key: $DEEPREAD_API_KEY" \
  -F "file=@contract.pdf" \
  -F "redaction_style=black_bar" \
  -F "language=en" \
  -F "webhook_url=https://your-app.com/webhooks/pii"

Response

Success Response (200 OK)

{
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "status": "queued"
}

Response Fields

Status Codes

GET /v1/pii/{job_id}

Retrieve the status and results of a PII redaction job.

Endpoint

GET https://api.deepread.tech/v1/pii/{job_id}

Authentication

X-API-Key: YOUR_API_KEY

Path Parameters

Example Request

curl https://api.deepread.tech/v1/pii/550e8400-e29b-41d4-a716-446655440000 \
  -H "X-API-Key: $DEEPREAD_API_KEY"

Response

Success Response (200 OK)

{
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "status": "completed",
  "progress_percent": 100,
  "redacted_file_url": "https://storage.deepread.tech/redacted/abc123.pdf",
  "report": {
    "id": "550e8400-e29b-41d4-a716-446655440000",
    "page_count": 3,
    "processing_time_ms": 4200,
    "total_redactions": 8,
    "redaction_policy": "black_bar",
    "confidence_threshold_used": 0.85,
    "pii_detected": {
      "EMAIL": {
        "count": 3,
        "pages": [1, 2],
        "confidence_avg": 0.97,
        "confidence_min": 0.91
      },
      "PHONE": {
        "count": 2,
        "pages": [1],
        "confidence_avg": 0.94,
        "confidence_min": 0.89
      },
      "SSN": {
        "count": 1,
        "pages": [2],
        "confidence_avg": 0.99,
        "confidence_min": 0.99
      },
      "NAME": {
        "count": 2,
        "pages": [1, 3],
        "confidence_avg": 0.91,
        "confidence_min": 0.86
      }
    }
  }
}

Response Fields

Report Fields

PII Detection Summary (per type)

Status Codes

Webhooks

When you provide a webhook_url, DeepRead sends a POST request when the job completes:

{
  "job_id": "550e8400-e29b-41d4-a716-446655440000",
  "status": "completed",
  "redacted_file_url": "https://storage.deepread.tech/redacted/abc123.pdf",
  "report": {
    "page_count": 3,
    "total_redactions": 8,
    "pii_detected": { ... }
  }
}

For failed jobs:

{
  "job_id": "550e8400-e29b-41d4-a716-446655440000",
  "status": "failed",
  "error": {
    "code": "DOCUMENT_CORRUPTED",
    "message": "Unable to parse the uploaded document"
  }
}

Webhook URLs must use HTTPS. Private IP addresses, localhost, and loopback addresses are rejected.

PII Types

The following PII types may appear as keys in the pii_detected report object:

Supported File Types

Rate Limits

Limits are per plan. Past a limit the request is refused with 429 and a Retry-After header saying when to try again.

A redaction job's pages count towards the account's pages in flight (16 on Free, 200 on Standard, 500 on Enterprise), shared with OCR and form-fill jobs. No document over 2,000 pages or 500 MB is accepted on any plan (413).

Response headers include:

• X-RateLimit-Limit: Maximum requests per period
• X-RateLimit-Remaining: Remaining requests
• X-RateLimit-Reset: Unix timestamp when limit resets

Error Codes