Developers
Document parsing API built for production
Upload documents, poll or webhook structured JSON with per-field confidence scores, and integrate parsing into your product without building OCR infrastructure.
Quick start
Create an API key in Settings → API keys in your workspace. Keys use the format dp_test_… (sandbox) or dp_live_… (production). The workspace is inferred from the key — no extra header required.
- Multipart upload with parser_id
- Poll result endpoint until status is complete
- Optional webhooks instead of polling
- Review queue for low-confidence fields
# Upload a document
curl -X POST https://api.autodocparse.com/v1/documents \
-H "Authorization: Bearer dp_test_your_key" \
-F "file=@invoice.pdf" \
-F "parser_id=YOUR_PARSER_UUID"
# Poll extraction result
curl https://api.autodocparse.com/v1/documents/DOC_UUID/result \
-H "Authorization: Bearer dp_test_your_key"Webhooks
Register HTTPS endpoints on your parser or pipeline. Payloads are signed with HMAC so you can verify authenticity before processing.
- document.completed — parsing finished, result available
- review.approved — human approved corrected fields
- document.failed — processing error
{
"event": "document.completed",
"data": {
"field_name": { "value": "…", "confidence": 0.98 }
},
"review_status": "approved"
}Authentication
Bearer token via API key, or session cookie from the dashboard. Rate limits apply per key and endpoint.
Export formats
JSON (default), CSV, XML, and XLSX from the dashboard or via export pipelines.
SDK
TypeScript and Python client SDKs are on the roadmap. Use the REST API directly until SDK packages are published.
Ship document parsing in your product
Create a workspace, generate an API key, and parse your first document in under five minutes.