- What it does — reads PDFs, images, and scanned documents; identifies the document type; and pulls out named fields with per-field confidence scores.
- Who it’s for — health-tech teams that need reliable, privacy-preserving extraction from clinical and administrative documents at scale.
- Where it runs — Azure Canada Central, Law 25-aligned, with optional de-identification.
How it works
POST /v1/documents call returns a 202 with a document id almost instantly. The structured result is ready seconds to minutes later — retrieve it by polling or by subscribing to a webhook.
How these docs are organised
Start here
Quickstart
Submit your first document and retrieve a structured result in under five minutes.
Authentication
Learn how API keys work and how to keep them safe.
Async model
Learn how to poll or subscribe to webhooks to retrieve your results.
Custom schemas
Define your own field list for any document type.