Document Agents
Your documents become queryable data
AI pipelines turn your PDFs, Word files, images and attachments into structured data: OCR extraction, automatic classification, progressive summaries and routing to the right destinations. All inside your sovereign perimeter.
Specifications
OCR & Extraction
PDF, Word, images into structured data
AI Classification
Automatic labeling and sorting
Hippocampe
L0/L1 summaries, progressive search
Routing
To the right destinations
Pipelines
9 types available
Destinations
Iceberg tables, object storage
Use cases
Sort incoming documents
Invoices, quotes, letters: pipelines automatically classify and file every incoming document
Automatic sorting, folder by folderQuery your PDFs
OCR extraction turns your documents into structured data, queryable like any other table
From PDF to Iceberg tableBuild a document memory
Hippocampe generates L0 and L1 summaries for progressive document search
Progressive searchIn action
The property manager
Karim manages a residential building and receives invoices, quotes and letters by email every day
- 1 Plug the email connector into the management inbox
- 2 Enable a document classification pipeline
- 3 Routing delivers each document to the right folder, property by property
- 4 Track runs from the pipelines dashboard
No more manual sorting: every document lands classified in the right place
The data lead
Léa needs to make years of accumulated PDF archives usable
- 1 Drop the archives into the organization's object storage
- 2 Run an OCR extraction pipeline on the batch
- 3 Structured data lands in Iceberg tables
- 4 Query the archives in SQL from the console
Years of archives become queryable data
Key benefits
Ready to make your documents talk?
Discover how document agents turn your documents into actionable data.
Request a demo