Document

Overview

The Document module provides a secure and easy way to convert PDFs and IDs into clean, machine‑readable data through FSCOs API.

Organizations integrate the Document AI service into their workflows to process both structured and unstructured documents, leveraging advanced AI models to extract, analyze and return standardized data that can be seamlessly incorporated into their business processes.

The Document AI service is also available as a SaaS service which can be white-labelled and rebranded. See the Hiiro Project for more information.

For a quick hands-on start, check out our @fsco/kickstart project on npm.

Key Features

• Template Management: Create and manage document templates to ensure consistent data extraction
• Using Collections: Create and manage collections to store your documents
• Document Processing: Convert PDFs and images into clean, machine‑readable data
• Workflows: Verify documents satisfy a complex set of related steps

How It Works

Template Creation

1. Create a template
2. Add fields to the template using a JSON schema or the template editor in the developer portal
3. Choose the AI provider and model to use for the template

Using Collections

1. Create a collection and optionally annotate it with tags
2. Add documents to the collection
3. Add templates to the collection to specify the data to be extracted from the documents as they are created for processing
4. View all documents in the collection

Document Processing

1. Create a document providing a document type
2. FSCO creates the document and returns a documentId
3. Upload a PDF or a image with a template identifier
4. When the document is processed, a webhook is triggered to notify you of the completion of the document processing
5. The same documentId can be used to retrieve the processed data

Common Use Cases

• Payables/Receivables: Extract data from invoices and receipts
• Compliance and Audit: Verify documents satisfy a complex set of related steps
• Insurance: Extract data from insurance documents
• HR: Extract data from employee onboarding documents
• Real Estate: Extract data from property documents
• Supply Chain: Extract data from purchase orders and invoices
• Other: Extract data from any document type

Getting Started

1. Create your first template
2. Create your first collection
3. Process your first document
4. View your documents

API Reference

For detailed API specifications, see the Documents API Reference.