Kapintelligence by Kapnative
AI DUE DILIGENCE

From data room to decision-grade review.

Assess funds, co-investments and single asset opportunities. Use standard DDQs and apply your own investment criteria, then build the recalculated model and IC memo - with analysts and managers validating before anything is final.

app.kapintelligence.com - fund DDQ run
Kapintelligence fund DDQ run: section completion by area, question cards with answers, attachments and sign-off states

A live fund DDQ run. Section completion on the left, answered questions with their attachments and sign-off state on the right.

First reviewable output in hours

Coverage, critical knock-out questions and first source-cited answers, subject to data-room scope.

Full institutional review: typically under 30 days

End-to-end process range, not software processing time. It depends on data-room completeness, manager response times and your own review steps.

Nothing is final without a person

Every draft carries a validation state: draft, manager-confirmed, analyst-approved or exception.

REVIEW DEPTHS

From quick screening to full due diligence.

Quick Screen

An initial assessment of fit, knock-out criteria and evidence coverage before committing resources to a full review.

Promise
Hours to first reviewable output, subject to data-room scope.

Full Review

A standard DDQ plus your own questionnaires, investment criteria and policy checks, with financial analysis, red flags and the IC memo.

Promise
Your configured institutional workflow, end to end.

Comparison / RFP

Compare shortlisted opportunities against the same criteria and the same methodology.

Promise
Decision support across opportunities, not one at a time.
THE WORKFLOW

Six steps, each with a visible state.

Select a step to see what the workspace produces at that point - and what your team gets out of it.

Step 1 of 6
Step 01

Assess coverage

Classify the data room against required documents, DDQ sections and IC evidence; identify missing or conflicting material.

Coverage map · 6 documents classified
COVERAGE SIMULATOR

See what a data room can and cannot answer - before you start reading.

Add or remove document types and watch which DDQ and IC requirements lose their evidence. This is the first thing the workspace does with a new data room.

Documents in the data room - select to add or remove
63%
Low
Coverage of 43 DDQ and IC requirements
Baseline manager facts Supported
Fund terms and investor rights Supported
Track record verification Missing
Fees and total cost ratio Partial
Valuation policy and NAV Partial
Risk and compliance framework Supported
Ownership and control Missing
Reporting and monitoring basis Supported
Generated manager request
4 requirements are unsupported
Deal-by-deal track record and audited statements
Reported performance cannot be recalculated without cash flows and audited figures.
Cost information sheet or PRIIPS KID
The total cost ratio cannot be reconciled to the fee schedule.
Latest quarterly report and valuation appendix
NAV and valuation methodology have no current source.
Firm structure and ownership chart
Ownership, control and conflicts cannot be mapped.

The workspace drafts the request list; the analyst decides what to send. Illustrative requirement set based on a European buyout DDQ.

Four unsupported requirements generate a drafted request list; the analyst decides what is sent.

What you get
Coverage map and a drafted manager request list
UNGATED

Read a full IC pack before you talk to us.

Anonymised and illustrative, with every important statement showing its validation status and evidence link - the same structure your committee would receive.

Open the sample pack
Executive summary Investment thesis Fund and manager facts Track-record analysis Portfolio construction Terms and economics Key risks Operational and legal findings Open questions Recommendation inputs Appendices and sources
GOVERNED, NOT JUST GENERATED

A writing assistant can draft. An institutional workflow must govern.

General-purpose models are useful, and analysts already use them. The difference is what a committee and an auditor can rely on afterwards.

Dimension General-purpose AI assistant Kapintelligence workflow
Sources
General-purpose AI assistant Whatever is pasted into the conversation
Kapintelligence workflow The approved data room, classified and version-controlled
Criteria
General-purpose AI assistant The prompt you happen to write
Kapintelligence workflow Standard DDQs plus your own criteria and policy checks, applied consistently
Calculations
General-purpose AI assistant Arithmetic inside a text response
Kapintelligence workflow Cash-flow model with visible inputs, methodology and reconciliation
Uncertainty
General-purpose AI assistant Rarely distinguishes missing from unknown
Kapintelligence workflow Missing, contradictory and awaiting-confirmation states are explicit
Validation
General-purpose AI assistant None
Kapintelligence workflow Manager confirmation, analyst review and correction history
Approvals
General-purpose AI assistant None
Kapintelligence workflow Named approval states on each output before it becomes final
Decision record
General-purpose AI assistant A chat log
Kapintelligence workflow An IC pack with sources, open questions and retained rationale
Data boundary
General-purpose AI assistant Depends on the consumer account and its terms
Kapintelligence workflow Tenant-isolated EU workspace under contractual no-training terms

A comparison of workflow properties, not of model quality. Kapintelligence uses large language models under contractual no-training terms; the governance around them is the product.

See Kapintelligence in action.

From fund due diligence to IC ready output.