document-processing-ocr

AI-powered procurement intelligence agent with local LLM, document OCR, vendor scoring, and web research. Analyze bids, extract structured data, and make decisions - 100% private.

Enterprise-grade document intelligence platform powered by a hybrid C++ Computer Vision engine and Agentic AI. Features local RAG, fraud detection, and GDPR-compliant PII redaction.

PaddleOCR 기반 OCR 검수·문서 변환·게시 워크플로우

Final BSc project developed by a two-person team: an intelligent document-management platform for organizing documents, extracted information, payments, guarantees, and deadlines.

Hybrid OCR system for extracting structured data from Nepali citizenship certificates. Uses local Tesseract preprocessing with Gemini Vision API fallback for optimal accuracy and privacy.

Four-agent trade document pipeline: OCR/LLM extraction, rule-based validation, compliance routing, NL-to-SQL queries.

AI-powered procurement intelligence agent with local LLM, document OCR, vendor scoring, and web research. Analyze bids, extract structured data, and make decisions - 100% private.

End-to-end document processing app (OCR + NER) built with Python, Tesseract, spaCy, and Streamlit.

Enterprise AI system to classify, split, and auto-route PDFs using Azure Document Intelligence and SharePoint.