IAM.Docs
Document ingestion, hybrid search and analysis with source citations.
Key capabilities
- PDF and office documents with OCR
- Hybrid search and source-grounded answers
- Contract and analytical document processing
Step-by-step guides
Use cases
Start with the outcome: open a guide, prepare prerequisites, follow the steps and verify the success signals.
01 Upload documents and get a cited answerDocuments are indexed and the answer includes verifiable source excerpts.
- Audience
- Analyst or legal professional
- Outcome
- Documents are indexed and the answer includes verifiable source excerpts.
Before you start
- Workspace and upload permission
- Supported PDF/DOCX/XLSX files
- Question and completeness criterion
Steps
- Create a collectionName the collection for the project and limit access to the required team.
- Upload filesWait for Ready; for OCR, verify language and recognition quality.
- Ask a questionSpecify period, documents and the expected response format.
- Verify citationsOpen every citation and verify page, excerpt and current file version.
Every material claim has a source and citations open the correct excerpt.
If citations are irrelevant, narrow collection/filter and verify OCR before rewriting the question.
02 Compare contract versions and collect risksChanges are grouped by section and every risk links to its source clause.
- Audience
- Legal or procurement analyst
- Outcome
- Changes are grouped by section and every risk links to its source clause.
Before you start
- Two clearly named versions
- Review template
- Risk taxonomy
Steps
- Upload both versionsAdd date/version metadata and exclude duplicate drafts.
- Run comparisonSelect baseline, new version and the appropriate review template.
- Review changesReview removed obligations, amounts, dates and liability first.
- Export the reportBefore export, open citations and mark confirmed or disputed items.
The report is reproducible from the same document ids and links to both versions.
For an incorrect diff, verify baseline/new order and OCR layout alignment.
Open detailed manual
Every application screen has a separate page with controls, safe example values, CLI/API alternatives and status-specific recovery steps.
Role in the ecosystem
IAM.Docs turns documents into a verifiable knowledge base. The system accepts PDF and office formats, extracts text and structure, builds search views and returns responses linked to pages and fragments of the source.
Main scenario
- Create a workspace and upload documents.
- Wait for processing, OCR and indexing.
- Search or ask a question about the collection.
- Check quotes in their original context.
- Save the report or export the structured result.
Search and evidence
Hybrid retrieval combines full-text and semantic signals. Answer without Sufficient evidence must report a limitation, not build on the fact. The quote stores the document ID, page/section and exact fragment, allowing you to reproduce the results after the session.
Access control
Documents, collections and search index are isolated by tenant/workspace. Deleting a source should cascade exclude it from retrieval. For protected request and response circuits can be inspected via IAM.Secure, and model routing is performed via IAM.Router.
Integrations
The web interface is suitable for an analyst, API is for downloading and batch workflows, MCP capability - for agent clients. IAM.Docs does not replace the system electronic document management and does not assert the legal significance of the conclusion.
Verified entry points
- Landing
- https://iamdocs.ru/en/
- Application
- https://app.iamdocs.ru/
- Portfolio
- https://iamgroup.ru/en/