Automate First Notice of Loss Intake with Custom AI

Most agencies rely on the built-in workflow tools in their Agency Management System. But rule engines in Applied Epic or Vertafore depend on structured data fields. They cannot parse the body of an email to decide if 'water damage in basement' is more urgent than 'fender bender in parking lot'. Adjusters still have to read every single email to assess severity.

This forces teams to use manual email forwarding and shared inboxes. A senior adjuster becomes a human router, spending 2-3 hours a day just reading and assigning new claims instead of managing complex files. This manual process is slow and error-prone, especially after hours or on weekends when staffing is low.

Consider an agency with 8 producers and 2 adjusters. An email arrives at 4:45 PM on a Friday titled 'Claim'. Inside is a two-paragraph description of a multi-car accident with potential injuries. Because the subject line is generic, it sits in the general claims inbox. The on-call adjuster does not see it until Monday morning, 64 hours after the First Notice of Loss. The client is already frustrated by the delay.

Syntora would initiate an engagement with a discovery phase, auditing your current FNOL intake channels, such as Microsoft 365 inboxes via Graph API and web form submissions via webhook. We would work with your team to define required data points for extraction and precise routing logic. A critical step involves analyzing a representative sample of your historical claim reports (e.g., 500-1000 documents) to develop and fine-tune prompts for the Claude API, ensuring accurate entity extraction and consistent JSON output across various report types.

The core of the system would be a custom FastAPI service, designed to orchestrate the entire workflow. This service would be triggered by an incoming FNOL report, often via an AWS Lambda function. The Lambda function would call the Claude API to parse the unstructured text, identifying key entities like policy number, claimant name, and a detailed incident description. A severity score, typically on a 1-5 scale, would then be assigned based on keywords and contextual understanding.

This structured data would then be integrated with your Agency Management System. Syntora would develop connectors using official APIs for platforms like Vertafore or Applied Epic, or utilize webhooks for other systems such as HawkSoft. The extracted claim summary, severity score, and recommended next steps would be written to a new activity or note within your AMS. Routing logic, customized to your agency's rules, would assign claims based on the AI-generated score; for example, directing higher-severity claims to senior personnel.

For transparency and control, every AI decision would be logged in a database, such as Supabase, alongside its confidence score. A configurable human review gate would be included for claims exceeding a predefined severity threshold. This would typically involve a notification to a designated channel (e.g., Slack) with a link to the claim and the AI-generated summary, allowing for a one-click approval before final assignment. This provides necessary human oversight for critical cases.