video1.0<iframe src="https://www.loom.com/embed/59e00be4807847b9be0b0a64124a10a9" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/59e00be4807847b9be0b0a64124a10a9-61020089f0be991a.gif115.889I showed our GitHub based application that is already running. From the web page, you can select a PDF and the related tests, then start the job. The app runs internally with OCR and PDF processing to extract relevant information, then highlights what matches the requirement and segments the output into 8 parts. It reviewed several cues and is currently around 90 percent accurate, with downloadable sample files you can inspect. I am looking to improve and improvise it further for the requirements.