Turn documents
into structured data
REST API that extracts clean JSON from any document — CVs, invoices, receipts, IDs, and more. Switch AI engines per request.
Supported documents
Everything you need
A complete extraction platform — from OCR to analytics, all accessible via a single REST API.
extracted fields
"name": "Jane Smith"
"role": "Engineer"
"exp": 5
"skills": [ … ]
Document Extraction
Extract structured fields from CVs, invoices, receipts, IDs, contracts, and any custom document — returned as clean JSON.
AI Translation
Translate extracted content into any language in the same API call. No separate translation pipeline needed.
Document Verification
Validate document authenticity, check field consistency, and detect anomalies — automatically flagged in the response.
processing…
200ms
Webhooks
Submit documents for async processing and receive HTTP callbacks the moment extraction completes — ideal for batch workflows.
Extractions this month
+24% vs last month
Analytics & History
Full audit trail of every extraction. Per-service usage dashboards, monthly trends, and token consumption — all in one place.
API Keys & Access
Create multiple API keys per account. Revoke, rotate, and scope access. Full token-based authentication out of the box.
Powered by best-in-class AI
State-of-the-art AI that understands documents the way humans do — handling complex layouts, inferring missing fields, and delivering industry-leading accuracy.
AI understands context, not just characters — extracts the right value even from ambiguous layouts.
Multi-column, tabular, and mixed-format documents handled automatically.
Missing or implicit fields are inferred from surrounding context.
Reads handwritten text and signatures with high confidence scores.
How it works
Three steps from document to structured data
Define a service
Configure a document service: choose a template that maps document fields to your JSON schema.
{
"service": "cv-extractor",
"fields": ["name", "skills",
"experience"]
}Send a document
POST a base64-encoded file to the extraction endpoint. Claude Sonnet processes the document and extracts structured fields.
{
"serviceCode": "cv-extractor",
"engine": "claude",
"fileBase64": "JVBERi0x..."
}Receive JSON
Get clean, structured JSON back in the response — or via webhook when using async mode.
{
"name": "Jane Smith",
"skills": ["React", "Node"],
"experience": "5 years",
"confidence": 0.98
}Start extracting today
Free tier includes 100 extractions per month. No credit card required. Upgrade any time.