A Generative AI Engineer has been asked to design an LLM-based application that accomplishes the following business objective: answer employee HR questions using HR PDF documentation. Which set of…

Question

Options

• ACalculate averaged embeddings for each HR document, compare embeddings to user query to
• BUse an LLM to summarize HR documentation. Provide summaries of documentation and user
• CCreate an interaction matrix of historical employee questions and HR documentation. Use ALS to
• DSplit HR documentation into chunks and embed into a vector store. Use the employee question to

• A
  16% (9)
• B
  3% (2)
• C
  9% (5)
• D
  72% (42)

Explanation

Option D describes the canonical RAG (Retrieval-Augmented Generation) pipeline, which is the standard architecture for Q&A over documents. HR PDFs are split into chunks, embedded into a vector store, and then the employee's question is used to retrieve the most relevant chunks, which are passed to an LLM to generate a grounded answer. Option A (averaging embeddings per document) loses granular detail. Option B (summarizing docs) loses specificity needed for precise HR answers. Option C (ALS/collaborative filtering) is a recommendation algorithm, not suitable for document Q&A.

No community discussion yet for this question.