VEERPAL SINGH

Insurance Doc Tracking- Cloud AI Architecture: Architected a scalable and cost-effective Azure cloud infrastructure to process over 100,000 documents daily, utilizing Azure Service, Azure Functions, PromptFlow, and Azure OpenAI. Document Classification Model: Developed a hierarchical, text-based classification system by fine-tuning transformer models like DistilRoBERTa. Advanced Data Extraction: Engineered a sophisticated data extraction pipeline for handling 100,000+ large, complex, structured, and unstructured documents per day using GPT-4.1, enhanced by an advanced Hybrid Retrieval-Augmented Generation (RAG). Signature Detection Methodology: Implemented a novel method for handwritten signature detection by correlating OCR text coordinates with computer vision analysis via OpenCV, achieving detection accuracy above 85%. RAG - Based Medical Chatbot: Architected an end-to-end Retrieval-Augmented Generation (RAG) medical chatbot integrating Gemini Flash for reasoning, Pinecone for vector indexing, and Sentence Transformers for semantic embeddings. LLM-as-a-Judge Pipeline for Automated Document Validation: Designed and implemented an end-to-end “LLM-as-a-Judge” validation pipeline to