Welcome
What will you learn in this course?
Who is the course for and prerequisites
© 2026 Dometrain. All rights reserved.
Build a Retrieval-Augmented Generation (RAG) chatbot that can answer questions using your data.
Retrieval-Augmented Generation (RAG) is a transformative AI architecture that enables large language models to answer questions using your specific data rather than relying solely on their training knowledge. It combines the power of semantic search through vector embeddings with the natural language capabilities of LLMs, creating AI systems that can provide accurate, contextual, and verifiable responses grounded in your custom knowledge base. RAG has become the cornerstone of modern AI applications, powering everything from intelligent customer support and internal knowledge bases to research assistants and domain-specific Q&A systems. Unlike traditional chatbots or pure LLM solutions, RAG-based systems can cite their sources, stay current with your latest data, and dramatically reduce hallucinations by anchoring responses in retrieved documents. Companies from startups to enterprises are adopting RAG to unlock the value in their documentation, support tickets, and proprietary content. In this hands-on course, instructor James Charlesworth will take you from understanding vector embeddings and semantic search to building a production-ready RAG chatbot in .NET with OpenAI, Pinecone, and advanced techniques like HYDE for enhanced retrieval accuracy.
The explanations are crystal clear and highly educational. James doesn't waste time and gets straight to the core concepts. It breaks down complex RAG concepts into manageable, easy-to-understand pieces. Highly recommended for any .NET developer looking to get into AI.
By the end of the course, I had a fully working chatbot querying my own data, exactly as promised.
The indexing and embeddings just right thing what I need know to continue my project.
Indexing techniques and RAG. Now I can extend the LLM knowledge with my project docs.
Real good explanation of how everything works. For my current assignment I had to build a RAG service in python and this was still extremely useful. It would be even better if the openai client was used and not all the MS extensions, but still a great course.
How the indexing and embedding really works. It helps to understand the foundation of rag.
James mixes the right amount of theory with hands on coding. Also the tool he builds is not boxed into one single use case, he show how to use with multiple data sources. This was awesome!
Retrieval techniques beyond cosine similarity, most other course don't explain it like James did.
The course helped me understand the key RAG concepts and how to implement them. While I wish it went a little deeper in some topics I think it provides an excellent opportunity for a Deep Dive course.
Creating the RAG pipeline, how to improve search results using HYDE and RRF/MMR, and tool implementation.
I loved how he thoroughly described the process and algorithms rather than skipping straight to the code.
I only had a vague understanding of how a semantic search worked. That was only one of many things I learned!
