ServiceXPro.ai case study
End-to-end rug cleaning workflow automation
From site visit to insurer-ready report
Challenge
Restoration teams needed one system for the whole claim: site capture, drying records, insurer reports, quotes and invoices.
What we delivered
Field capture and floor plans
- iOS app in React Native and Expo, sharing one NestJS backend with the web app.
- Initial Assessment with 16 sections and 161 fields, saved and resumed on site.
- LiDAR room and structure scans through a native Apple RoomPlan module.
- Photos, video and files stored per job, room and chamber.
Drying, reports and finance
- Structures, drying chambers and rooms laid out for every job.
- Moisture points by meter and mode, tracked against the dry standard.
- IICRC chamber equipment calculator and equipment tracking.
- Project and Initial Assessment reports as PDF, with a report builder.
- Quotes, estimates and invoices, with finance limited to owners and admins.
AI and production platform
- “Tidy text” on Amazon Bedrock (Claude Haiku 4.5) cleans up technician notes.
- Further AI agents set up on Amazon Bedrock, alongside Tidy text.
- Ten automatic checks keep numbers, dates, names and negations unchanged.
- The user accepts every suggestion; note text is never stored or logged.
- Multi-tenant organizations with five roles and a platform admin.
- DevOps on AWS: staging first, then production releases gated on health checks, with alerting.