Extract company mentions
Identify names, identifiers, countries and document context during ingestion.
COMPANY DATA FOR RAG
Build entity-aware retrieval for enterprise AI.
Ground the answer before action.
Enrich documents and retrieval indexes with canonical company identifiers so enterprise AI retrieves evidence for the correct legal entity.
START WITHDocuments, company mentions and metadataREVIEWEntity-linked records + retrieval filters
Design the agent around evidence and policy.Define the agent host, tool scope, countries, evidence fields, exceptions and permitted actions.
WHEN TO USE IT
Prevent retrieval systems from mixing documents about similarly named companies, subsidiaries or group brands.
Give the agent a structured entity result, provenance and explicit exception state. Keep the evidence it relied upon beside the answer or action.
- A RAG system indexes contracts, filings, supplier records or company research.
- Search results confuse a parent company, subsidiary, branch or trading brand.
- Enterprise AI needs jurisdiction and identifier filters for reliable retrieval.
THE WORKFLOW
Four steps. One clear result boundary.
Resolve each entity
Match mentions to canonical legal-company records and preserve unresolved or ambiguous states.
Enrich the index
Attach registration number, tax ID, jurisdiction, parent context, source and timestamps as metadata.
Retrieve with entity filters
Use stable identifiers and jurisdiction context to constrain the evidence supplied to the model.
WHAT YOU CAN REVIEW
The record, not just the number.
Fields vary by country and company. VATFind labels absent information instead of inventing a result.
- 01Canonical company ID
- A stable key for connecting documents and records that refer to the same legal entity.
- 02Entity aliases
- Legal and submitted names retained without treating every alias as a separate company.
- 03Jurisdiction metadata
- Country, region and identifier types used as retrieval filters.
- 04Company attributes
- Selected identifiers, status and address fields for enrichment and disambiguation.
- 05Provenance metadata
- Source and retrieval time attached to the entity-resolution result.
READ THE RESULT CORRECTLY
Vector similarity is not company identity.
Embeddings can find semantically similar text but do not reliably distinguish legal entities. Entity resolution should create the identity layer, while the RAG system retains document-level permissions, citations, freshness and relevance controls.
Review data and limitationsCOMMON QUESTIONS
Before you implement the workflow.
These answers define how the agent should use company evidence, handle ambiguity and stay inside policy.