Understand the RAG pipeline
Trace a question through retrieval and generation, and see what each stage contributes.
- Describe the stages in a basic retrieval-augmented generation system.
Retrieval-Augmented Generation (RAG) gives a language model relevant external material at answer time. A typical flow receives a question, retrieves useful passages from a prepared collection, then asks the model to answer using those passages. Retrieval can make answers more relevant to a private or changing knowledge base, but it does not guarantee truth.
A small example
1question = "When does the library close?"
2passages = ["Hours: open until 8 pm", "Address: 10 Main Street"]
3context = passages[0]
4print(f"Question: {question}\nContext: {context}")Question: When does the library close? Context: Hours: open until 8 pm
RAG has two broad phases: prepare an index from source material, then retrieve and generate for each query. The model only sees the context actually provided in its prompt. Missing, stale, or irrelevant context can still lead to a poor answer.
Key takeaways
Describe the stages in a basic retrieval-augmented generation system.
Check that retrieved evidence is relevant, current, and allowed for this user.
Lesson quiz
5 questions · pass with 4 correct · up to 50 XP
Passing this quiz completes the lesson and keeps your streak going. Questions you miss come back in review sessions later.
Questions about this lesson
Stuck? Ask. Figured something out? Share it. Explaining is one of the best ways to learn.
Loading posts…