I analyzed a sample set of pharmaceutical vendor documents to identify their structure and challenges for AI extraction.

I prepared and processed pharmaceutical documents using Python, covering data cleaning, JSON flattening, and image preprocessing to improve OCR readiness.

I extracted and structured text and bounding-box data from multi-page SDF PDFs using PyMuPDF, then analyzed extraction quirks and proposed regex+location heuristics for field linking.

I compared Tesseract, PaddleOCR, and EasyOCR on a scanned pharmaceutical supplier PDF and produced a written analysis plus Colab code.

I built and tested a RAG pipeline for document retrieval using LlamaIndex, implementing hybrid retrieval, embeddings, chunking, and reranking.

I compared three RAG configurations on a sample SDF document, logged retrieved chunks and answers, and analyzed which settings produced the most precise answers.

I implemented page-level segmentation and metadata tagging to split a bundled pharmaceutical PDF into individual documents and label each page by type.

I built a Gradio-based document Q&A chatbot interface and integrated it with a retrieval pipeline for PDF question answering.