CortiX DocIntel

Turns PDFs, Word, Excel, PowerPoint, CSV, text and images into clean, typed data with the source line for every value. Built-in templates or your own fields, review before export and an API for back offices.

What it does

CortiX DocIntel turns PDFs, Word, Excel, PowerPoint, CSV, text and images into clean, typed data with a source line for every value.

Templates: pick a built-in template (invoice, receipt, certificate of insurance, purchase order, bank statement, business card, key facts) or define your own fields and types.

Jobs: upload one file or many; each file becomes a document with its extracted fields.

Citations: every value carries the line it came from. A value the document does not show is marked not found. A value with no source is marked unverified.

Scans: Amazon Textract OCR reads scanned pages.

Review: people correct values before export and their corrections are kept.

Exports: JSON, CSV, XLSX and a ZIP of all.

API: /api/v1/docintel/v1 with a CortiX API key, for developers and back offices.

The monthly plan includes 2,000 pages a month. Your data stays in your CortiX workspace and never crosses workspaces.

Highlights

Typed fields with citations: every value shows the line it came from; values the document does not show are marked not found.
Built-in templates for invoices, receipts, certificates of insurance, purchase orders, bank statements and business cards. You can also define your own fields.
People review values before export to JSON, CSV or XLSX; developers use the same engine through an API key.