Service description

TabInt helps private-markets teams turn fragmented documents into structured, source-linked deal records they can review, search, and follow over time.

Designed for document-based workflows, including private equity and private credit, the platform organizes information around deals, their source documents, and structured sheets of extracted attributes. TabInt organizes and presents information; it does not make or recommend investment, credit, valuation, or other deal decisions.

How it works

1. Bring in and organize deal documents

Customers can upload files, connect SharePoint, or import from an IMAP mailbox.

For larger collections, TabInt can suggest how to group and organize documents. Users review and adjust the suggested structures.

2. Define the information to extract

Customers can provide a schema, a defined set of fields, or derive one from a document, then apply it across one or more files. TabInt can also suggest schemas or apply customer-selected standard schemas.

TabInt extracts values and dates into structured sheets. Each extracted value includes a confidence score.

3. Review values and inspect their sources

Users can approve, reject, edit, or rerun individual extracted values. Selecting a value opens its source document at the relevant page or highlighted text so the user can manually confirm it.

Confidence scores help users prioritize review; they do not guarantee accuracy. AI-generated extractions and chat responses may be inaccurate or incomplete. Users should verify material information against source documents before relying on it.

4. Search and ask questions

Users can search across their documents and ask questions about one or more deals through chat. They can also review and adjust the extraction workflow through the chat interface.

5. Recreate historical reports

Users can recreate and export a report for a past date using either the information available then or what they know today. This distinguishes valid time, when a fact was true, from transaction time, when it was recorded in TabInt, and shows how later updates changed the picture.

The timeline also records decision time, when decisions about facts were made. Users can navigate this history and trace values to their sources, but cannot edit the time dimensions through the timeline control.

6. Export structured information

Customers can download structured data and recreated historical reports as CSV, Excel, or JSON.

Supported content and integrations

TabInt supports English-language PDF, Word, and Excel files, email content, and attachments. It processes both scanned and text-based documents using optical character recognition and AI-assisted document processing.

SharePoint

SharePoint integration performs incremental sync of content the connecting user can access. Synced content is copied into TabInt and is initially visible only to that user. The user can then share it with other TabInt users.

IMAP email

Users choose whether to import attachments only or attachments together with email content. Imports copy the selected material into TabInt.

Other data connectors

Additional data connectors, including OneDrive, are available upon request. Connector availability, configuration, and scope are agreed with each customer and documented in the applicable proposal or order form.

Security and hosting

TabInt's primary platform runs on Google Cloud in the United States. All customer-data hosting, including backups, is located in the United States.

Customer environments are tenant-isolated, and data is encrypted in transit and at rest. TabInt supports single sign-on and audit logs.

TabInt has started the SOC 2 Type II process. A completed report is not yet available. View the TabInt Trust Center compliance overview or contact security@tabint.ai.

AI providers and processing

TabInt uses OpenAI and Anthropic services for AI-assisted extraction, schema suggestions, and chat. Relevant customer content may be sent to these providers to perform those operations.

Provider processing and retention depend on the service and configuration used. For current information about provider processing, retention, and use of customer content, contact security@tabint.ai.

Data retention and deletion

Customer data is retained while the customer account is in good standing. After account termination, data is retained for 30 days to allow reinstatement, then deleted within a further 30 days.

Customers may request deletion at privacy@tabint.ai. Data that Tabular Intelligence has no legal or business obligation to retain is deleted within 30 days of the request.

Deleted data may remain in encrypted backups until those backups expire, up to 14 days after deletion from production systems. See the Privacy Policy for more information.

Deployment and commercial terms

TabInt does not publish universal file-size or processing-volume limits. Suitability, implementation, onboarding, support, service scope, and fees are defined for each customer in its proposal and order form.

Contact sales@tabint.ai to discuss a deployment.

This service description explains the service generally. The applicable customer proposal, order form, and agreements define the customer's specific scope and commercial terms.