https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/Recent content in Retrieval-Augmented Generation on DATATWEETSHugoenCopyright (c) 2025 DatatweetsSat, 27 Jun 2026 09:00:00 +0200https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-1-what-rag-is/Fri, 14 Nov 2025 09:00:00 +0200https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-1-what-rag-is/RAG gives a language model access to your own knowledge by retrieving relevant documents and putting them in the prompt. See the retrieve-then-generate pattern run for real, and learn why it keeps answers current, private, and grounded.https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-2-building-a-rag-pipeline/Fri, 14 Nov 2025 09:00:00 +0200https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-2-building-a-rag-pipeline/Build a real RAG pipeline as composable Python: retrieve relevant documents from Chroma, assemble a grounded prompt, and generate an answer with Claude — wrapped into a single reusable answer() function you can point at any knowledge base.https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-3-chunking-documents/Fri, 14 Nov 2025 09:00:00 +0200https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-3-chunking-documents/A whole document is too much for one embedding. Learn why chunking matters, write a fixed-size chunker with overlap, store chunks in Chroma with metadata, and retrieve the single most relevant chunk — the step that makes RAG precise.https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-4-grounding-and-citations/Fri, 14 Nov 2025 09:00:00 +0200https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-4-grounding-and-citations/Turn a working pipeline into a trustworthy one. Force the model to answer only from retrieved context, decline when the data can’t support an answer, and cite numbered sources — the difference between a demo and a product.https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-5-guided-project-docs-qa-bot/Fri, 14 Nov 2025 09:00:00 +0200https://datatweets.com/courses/generative-ai/retrieval-augmented-generation/lesson-5-guided-project-docs-qa-bot/The capstone: combine chunking, a persistent vector database, grounding, and citations into a documentation Q&A bot. Load a real handbook, embed it once, and ask() questions to get cited answers — and watch it decline what the docs don’t cover.