Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn to build a multimodal intelligent document processing pipeline using Nemotron RAG in this 59-minute developer-focused livestream from Nvidia. Discover how to create an AI agent that can answer questions about complex PDFs containing dense tables, charts, and scanned pages while providing precise, cited snippets from original documents without writing brittle parsers. Follow along with a live demonstration that shows how to build and run an intelligent document processing agent using GPU-accelerated extraction and just a few hundred lines of Python code. Explore the integration of NeMo Retriever library and Nemotron RAG models to structure text, tables as markdown, and chart crops for high-recall multimodal retrieval. Master the techniques for tuning chunking, embeddings, and reranking to ensure reliable surfacing of relevant pages and tables that answer complex queries. Learn to wire top results into a Nemotron-powered assistant that delivers grounded, traceable answers with citations to exact pages, tables, or figures. Walk away with a complete understanding of how Nemotron RAG retrieves information from documents and a reusable Python pipeline ready for adaptation to your own intelligent document processing workloads.
