# Subrogation opportunity agent

> Find paid claims where another party may be liable, and prepare the recovery file. Recoveries are missed when liability clues are buried in notes and documents.

Source: https://datanature.ai/use-cases/subrogation-opportunity-agent-insurance
Publisher: Dlytica Inc. (DataNature is a Dlytica company)
Updated: 2026-11-05

Industry: Insurance. Function: Finance and reporting. Built with: Voir, DataNature, DocuContext, AI360 Studio.

## What does the subrogation opportunity agent do?

Recoveries are missed when liability clues are buried in notes and documents. A Voir agent reviews claim notes, police reports and invoices for third-party liability indicators, ranks opportunities by expected recovery and drafts the demand package for the recovery team.

## How is the agent designed?

- **Pattern**: Single agent with document retrieval over claim files and SQL over certified claims cubes, plus a scoring model for recovery likelihood.
- **Data scope**: Closed and open claims in the recovery team's lines, with claimant personal data masked except for authorized staff.
- **Human approval**: Sending a demand or contacting another insurer requires recovery specialist approval.
- **Platform components**: Voir, DocuContext, AI360 Studio, Semantic Layer, Atlas

## How does it work on DataNature?

1. **Ground it in governed data.** Certified Gold tables and semantic-layer cubes give the agent consistent numbers; DocuContext partitions give it documents with links to source; the knowledge graph gives it relationships between entities where the use case needs them.
2. **Scope the twin.** The Voir twin's data scope, document partitions and SQL allowlist are enforced by the Agent Meta-Harness inside every tool call, and Atlas applies masks and row filters for the user it acts for.
3. **Route the model.** The AI gateway sends requests to an approved model endpoint and logs tokens, cost and latency.
4. **Keep a human in the loop.** The agent recommends; people approve actions that change systems.
5. **Trace and evaluate.** Every turn is traced, and answer quality is evaluated against real questions before and after each change.

## What data does it use?

- Claims, payments and reserves
- Police and incident reports
- Claim notes
- Recovery history

## Why run it on a sovereign platform?

The data, prompts, retrieved passages, model inference and logs all stay inside your jurisdiction, so the agent can work on regulated data without sending it to a foreign service.

## How do we get started?

Dlytica Consulting can scope and build it, NeuralShore engineers can join your team, and Dlytica Academy-trained talent is ready to support it. Book a demo with Dlytica to see it on your own data.
