AIF-C01 Practice Question: Building a RAG solution using Amazon Bedrock…

A company is building a RAG solution using Amazon Bedrock Knowledge Bases. Which TWO steps are essential in the document ingestion pipeline? (Select TWO.)

⚠ Common exam trap

AWS often tests the distinction between the ingestion pipeline (chunking and embedding) and the inference pipeline (model endpoints, guardrails, and fine-tuning), so candidates mistakenly select options that belong to the query or model customization phase.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Generating embeddings for each chunk

Option D is correct because chunking documents into smaller segments is a required preprocessing step in a Bedrock Knowledge Bases ingestion pipeline; splitting source documents into manageable chunks allows the system to process and index content effectively. Option C is correct because generating embeddings for each chunk is essential: Bedrock invokes an embedding model (for example, Amazon Titan Embeddings or Cohere Embed) to convert each chunk into a vector, which is then stored in the vector store for semantic retrieval during RAG queries. Option A is not essential to the ingestion pipeline because a model endpoint for real-time inference relates to query-time generation, not document ingestion. Option B is incorrect because Bedrock Guardrails apply content filtering and safety policies at inference time and are not a required ingestion step. Option E is incorrect because fine-tuning the model on the documents is a separate model-customization activity and is not part of the Knowledge Bases ingestion pipeline, which relies on embeddings and vector storage rather than fine-tuning.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Setting up a model endpoint for real-time inference
Why it's wrong here
A real-time inference endpoint serves queries after the knowledge base is built; ingestion itself only chunks, embeds and writes vectors. Endpoints are correct for serving retrieval-augmented responses to users. The pipeline's essential steps are parsing source documents and generating embeddings stored in the vector index.
✗
Creating a Bedrock Guardrail for the documents
Why it's wrong here
Guardrails filter model inputs and outputs at inference time; they do not parse, chunk or embed source documents during ingestion. Guardrails are correct when enforcing content safety and denied topics in a deployed application. Ingestion instead requires chunking and embedding into a vector store.
✓
Generating embeddings for each chunk
Why this is correct
Embeddings convert each chunk into vectors that capture semantic meaning, enabling similarity search against the user query. Without this vectorisation step, Amazon Bedrock Knowledge Bases cannot retrieve relevant passages, so the RAG pipeline fails at its core retrieval stage.
✓
Chunking documents into smaller segments
Why this is correct
Chunking splits source documents into smaller segments before embedding, because embedding models have fixed token limits and retrieval precision improves with focused passages. This segmentation is a mandatory ingestion stage in Amazon Bedrock Knowledge Bases, feeding each chunk to the embedding model.
✗
Fine-tuning the model on the documents
Why it's wrong here
Fine-tuning alters model weights and is unrelated to ingesting documents into a knowledge base; retrieval supplies grounding instead. Fine-tuning suits teaching task-specific behaviour or style. Ingestion requires chunking documents and converting them to embeddings written into the vector store.

Go deeper

AIF-C01 Practice Question: Building a RAG solution using Amazon Bedrock…

Why each option matters

Option-by-option breakdown

Related to this question