Building RAG Agents with LLMs
NVIDIA Deep Learning Institute
Last updated August 15, 2026
Building RAG Agents with LLMs is an eight-hour, hands-on course from NVIDIA's Deep Learning Institute for developers who want to build and deploy retrieval-augmented LLM agents in practice. Working in Python with tools like LangChain, FAISS, and LangServe, you design a system that holds a conversation, reasons over documents, manages dialog state, and answers questions from its own data — then evaluates and scales it. A practical, code-first path into the engineering behind production RAG agents.
What you'll learn
- Designing an LLM system that reasons with internal and external tools
- Managing dialog state and turning documents into structured information
- Using embeddings for similarity search and guardrailing
- Building, modularizing, and evaluating a working RAG agent
Frequently asked questions about Building RAG Agents with LLMs
Who is Building RAG Agents with LLMs for?
Developers with intermediate Python and some deep-learning background who want to build real RAG agent systems.
Is Building RAG Agents with LLMs free?
No — Building RAG Agents with LLMs is a paid course.
What are the prerequisites for Building RAG Agents with LLMs?
Introductory deep learning knowledge (comfort with PyTorch and transfer learning preferred) and intermediate Python, including object-oriented programming and libraries.
Do you need to code for Building RAG Agents with LLMs?
Yes — Building RAG Agents with LLMs involves hands-on coding.
Why we suggest this course
Most RAG material stops at the concept; this one has you build and deploy an agent end to end. For developers ready to move from understanding RAG to shipping it.