video1.0<iframe src="https://www.loom.com/embed/e8d88ea52975413ba63120ed0bb33765" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/e8d88ea52975413ba63120ed0bb33765-3ba20ce910068452.gif194.714This Loom demonstrates QRI’s platform for automating download accounting by extracting security and transactional data from uploaded voice and document files. It addresses a manual workflow where employees extract totals and copy dates into accounting systems, which is slow and error prone, and aims to scale fully automated processing. The multi-tenant system uses Google (GWT) authentication with multi-factor authentication and role-based access control, built with React on the front end and postgreSQL on the back end. Uploaded documents move through a two-stage processing pipeline, with approvals and optional editing, and the system completes and stores results in a database and supports export. The author notes processing is slower on localhost but should be 3 to 5 times faster when deployed on a server.