Document-aware
Identifies invoices, receipts, IDs, forms, contracts, and custom document types before extraction begins.
01 Document intelligence infrastructure
Production environment
SCE identifies what a document is, extracts what matters, resolves ambiguity, and returns data your applications can trust.
01ClassifyUnderstands document type
02ExtractMaps data to your schema
03ResolveAsks only when uncertain
02 Capture framework
Version 0.1 accepts, fingerprints, preserves, records, and queues each source. Document understanding is intentionally reserved for a later version.
Version 0.1 returns capture_id, status, and durable storage_location.
03 Canonical pipeline
Every capture receives the same ordered workflow. Version 0.1 completes intake and preservation, then stops at the Image Processing boundary.
OCR, classification, AI, rules, human confirmation, JSON output, and project integration are framework stages only. No processor is implemented yet.
04 More than OCR
Identifies invoices, receipts, IDs, forms, contracts, and custom document types before extraction begins.
Normalizes every result to a versioned schema, so downstream applications receive consistent data.
Surfaces focused follow-up questions when confidence is low instead of guessing or failing silently.
Stores the original file and maintains a traceable relationship between source, extraction, and corrections.
05 Independent plugins
Product integrations live outside SCE as separately installed plugins. Each plugin uses the same versioned contract without adding application knowledge to the core.
curl -X POST https://capture.combridgecorp.com/api/v1/capture \
-H "Authorization: Bearer $SCE_API_KEY" \
-F "file=@invoice.pdf" \
-F "schema_id=invoice.v1"