Extract structured data from any document

Invoices, receipts, resumes — send any document, get clean JSON back. See real extraction results below.

Invoice Extraction

Turn invoices into structured line items

Extract vendor name, invoice number, dates, line items with descriptions and amounts, tax, and totals — all with confidence scores. Works with any invoice layout.

📄 Input — invoice_2024.pdfBefore
INVOICE #INV-2024-0847
From: Acme CorpDate: 2024-03-15
To: TechStart Inc.Due: 2024-04-15
Cloud hosting (March)$1,200.00
API access — Pro tier$499.00
Support hours (8h × $150)$1,200.00
Subtotal$2,899.00
Tax (8.5%)$246.42
Total Due$3,145.42
✨ Output — Extracted JSONAfter — 1.2s
{ "vendor": "Acme Corp", "invoice_number": "INV-2024-0847", "date": "2024-03-15", "due_date": "2024-04-15", "line_items": [ { "description": "Cloud hosting (March)", "amount": 1200.00 }, { "description": "API access — Pro tier", "amount": 499.00 }, { "description": "Support hours (8h × $150)", "amount": 1200.00 } ], "subtotal": 2899.00, "tax": 246.42, "total": 3145.42, "confidence": 0.97 }
curl -X POST https://docuextract.dev/v1/extract \
-H "Authorization: Bearer dk_live_..." \
-H "Content-Type: application/json" \
-d '{"document": "https://example.com/invoice.pdf", "type": "invoice"}'
Try invoice extraction in the playground →
Receipt Processing

Parse receipts into expense data

Extract merchant, date, items, tax, tip, and total from any receipt — even crumpled photos. Perfect for expense tracking, bookkeeping automation, and travel reimbursement.

📷 Input — receipt_photo.jpgBefore
BLUE BOTTLE COFFEE
123 Market St, San Francisco CA
Cortado$5.50
Avocado Toast$14.00
Sparkling Water$3.00
Subtotal$22.50
Tax (8.625%)$1.94
Tip$4.50
Total$28.94
03/15/2024 11:42 AM
✨ Output — Extracted JSONAfter — 0.9s
{ "merchant": "Blue Bottle Coffee", "address": "123 Market St, San Francisco CA", "date": "2024-03-15", "time": "11:42", "items": [ { "name": "Cortado", "price": 5.50 }, { "name": "Avocado Toast", "price": 14.00 }, { "name": "Sparkling Water", "price": 3.00 } ], "subtotal": 22.50, "tax": 1.94, "tip": 4.50, "total": 28.94, "confidence": 0.94 }
curl -X POST https://docuextract.dev/v1/extract \
-H "Authorization: Bearer dk_live_..." \
-d '{"document": "<base64_receipt_image>", "type": "receipt"}'
Try receipt extraction in the playground →
Resume Parsing

Convert resumes into candidate profiles

Extract name, contact info, skills, work experience, and education from PDF resumes. Build ATS integrations, talent databases, or hiring automation in minutes.

📄 Input — sarah_chen_resume.pdfBefore
Sarah Chen
sarah@example.com · San Francisco, CA · github.com/schen
Experience
Senior Software Engineer — Stripe (2021–Present)
Led payments infrastructure team. Reduced checkout latency by 40%.
Software Engineer — Vercel (2019–2021)
Built Edge Functions runtime. 12 open-source contributions.
Skills
TypeScript, Go, Python, PostgreSQL, Redis, AWS, Kubernetes
Education
B.S. Computer Science — UC Berkeley (2019)
✨ Output — Extracted JSONAfter — 1.5s
{ "name": "Sarah Chen", "email": "sarah@example.com", "location": "San Francisco, CA", "experience": [ { "title": "Senior Software Engineer", "company": "Stripe", "period": "2021-present", "highlights": ["Led payments infrastructure"] }, { "title": "Software Engineer", "company": "Vercel", "period": "2019-2021" } ], "skills": ["TypeScript", "Go", "Python", "PostgreSQL"], "education": [{ "degree": "B.S. Computer Science", "school": "UC Berkeley", "year": 2019 }], "confidence": 0.95 }
curl -X POST https://docuextract.dev/v1/extract \
-H "Authorization: Bearer dk_live_..." \
-d '{"document": "https://example.com/resume.pdf", "type": "resume"}'
Try resume extraction in the playground →

8 document types, one endpoint

DocuExtract auto-detects the document type, or you can specify it for higher accuracy.

📄
Invoices
🧾
Receipts
📋
Contracts
📝
Forms
💳
Bank Statements
📎
Resumes
🪪
ID Documents
📇
Business Cards

Extract your first document in 3 minutes

100 free extractions/month. No credit card. No templates. No training.

Try the playground →Read the docs