This project is designed to operate at the intersection of image processing, Optical Character Recognition (OCR), and deep learning, incorporating both traditional and modern AI techniques. The system will accept input in the form of either a PDF document or an image. The first step of the pipeline will involve the extraction of all textual content using Tesseract OCR, which will be responsible for converting visual character data into machine-readable text. In parallel, the system will utilize deep learning models, particularly Convolutional Neural Networks (CNNs) and YOLO (You Only Look Once), to perform layout analysis, region detection, and possibly even object or form-field recognition within the document. These models will help identify key sections, structures, or entities that can guide how the extracted text is further interpreted and organized. Once the raw textual and structural information has been obtained, it will be processed by a Large Language Model (LLM). The LLM will be responsible for interpreting the content contextually, extracting semantically relevant information, and transforming it into a structured tabular format that is both user-friendly and ready for downstream applications. The final structured data will then be stored in a database, allowing for efficient retrieval, indexing, and future querying. By combining OCR, CNN/YOLO-based visual understanding, and LLM-driven text interpretation, this project aims to deliver a powerful end-to-end solution for automated document intelligence. The system has broad application potential in fields such as invoice processing, legal document analysis, academic archiving, and beyond.