This Project is under development as per demand from a client. This system allows its users to single upload or bulk upload buyer and seller documents. The backend AI is trained via prompt engineering to evaluate seller documents based on 'n' number of evaluation criteria as per client requirements. This system is designed to read unformatted scanned PDF documents using advanced AI OCR engine. It is capable to read low quality PDF document pages which are blurred, having stains, scratches, folds, etc which is very much difficult to be read by standard OCRs. The system can also identify ink seals and signatures in document Python backend component is responsible to breakdown seller's each documents into 'n' number of pages and provides each page one by one to AI for accurate and successful evaluation. System is capable to identify and reduce size of too large document pages causing timeouts. Multiple seller documents can be selected for evaluation at once. System implements a queue mechanism to process each seller documents one-by-one and shows the evaluation progress anytime whenever the user wants. System also has a parallel page processing mechanism which significantly increases the evaluation speed. In case if some pages of seller document is rotated which causes inaccuracies in reading it, then system allows users to see, rotate and save such pages before evaluation. System lists all the evaluated seller documents with details of each. The details shows the list of all evaluation criteria, whether its qualified or not, document name and page number where its evaluation contents are found, reason for evaluation contents to exists, etc. The system is designed to run either on off-premises cloud infrastructure or on-premises LAN infrastructure. The backend is robustly designed to ensure data confidentiality by localizing backend operations and using cloud services temporarily with complete cleanup after its job is completed.