video1.0<iframe src="https://www.loom.com/embed/31eb711cc7464bcc8ce8ddb64d62d85b" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/31eb711cc7464bcc8ce8ddb64d62d85b-5f993f63c6835462.gif381.068In this video, I present an AI-powered claim processing pipeline I built using FastAPI and LangRaph that automates the classification and extraction of data from various insurance claim documents. The system utilizes a segregator agent to classify pages into nine document types, which are then processed by specialized agents for identity, discharge, and billing information. I implemented fallback logic to ensure no pages are missed and normalized outputs to prevent errors. The demo showcases the structured JSON response with extracted patient details, discharge information, and billing data totaling 6480. I encourage viewers to consider how this scalable solution can be extended with advanced features like Vector Database integration for handling larger claim volumes.