A civic-garden hotline already has a plot-rules index built by another team. The chatbot must fetch the nearest three blurbs for watering on a holiday. What should the application code call?
Select an answer to reveal the explanation.
Short Explanation
The plot-rules index already exists. The chatbot should query the vector-database client for top-k hits and take the texts and scores. Do not rebuild ingest on every greeting, redefine RAG in a comment, or scan a spreadsheet by hand.
Full Explanation
At application depth, retrieval is a client call: query the existing vector index for top-k texts and scores, then hand those hits to the next step. Ingest, chunking, and index design belong to a later data-pipeline domain, and restating the RAG definition is not a client call. Rebuilding the store on every greeting, or scanning a spreadsheet by hand, is not how the chatbot uses the index.