A company wants to fine-tune a foundation model (FM) for a specific use case. The company needs to deploy the FM on Amazon Bedrock for internal use.
Which solution will meet these requirements?
A. Run responses that have been generated by a pre-trained FM through Amazon Bedrock Guardrails to create the custom FM.
B. Use Amazon Personalize to customize the FM with custom data.
C. Use conversational builder for Amazon Bedrock Agents to create the custom model.
D. Use Amazon SageMaker AI to customize the FM. Then, import the trained model into Amazon Bedrock.
Show Answer
Correct Answer: D
Explanation: Amazon SageMaker AI supports fine-tuning and customizing foundation models. After customization, the model can be imported into Amazon Bedrock using the custom model import capability for managed deployment and internal use. Guardrails do not create or fine-tune models, Amazon Personalize is for recommendation systems, and Bedrock Agents orchestrate workflows rather than train foundation models.
Question 95
Which task describes a use case for intelligent document processing (IDP)?
A. Predict fraudulent transactions
B. Personalize product offerings
C. Analyze user feedback and perform sentiment analysis
D. Automatically extract and format data from scanned files
Show Answer
Correct Answer: D
Explanation: Intelligent Document Processing (IDP) is designed to extract, classify, and structure information from scanned, unstructured, or semi-structured documents such as invoices, forms, receipts, and PDFs. Predicting fraud, personalizing offers, and sentiment analysis are different AI/ML use cases.
Question 96
A company is using Amazon Q Business to create an AI assistant. The company needs to restrict user interactions with the AI assistant to company-approved topics.
Which feature will meet these requirements?
A. Amazon Q Business Enterprise index
B. Amazon Q Business Starter index
C. Amazon Q Business application guardrails
D. Amazon Q index cross-account access
Show Answer
Correct Answer: C
Explanation: Amazon Q Business application guardrails allow organizations to restrict assistant behavior and responses to approved topics, enforce content boundaries, and align interactions with organizational policies. Index types (Starter or Enterprise) relate to indexing capacity and features, while cross-account access concerns permissions rather than topic restriction.
Question 97
HOTSPOT
-
A company wants to build a new ML solution. The company already has data. The company needs to understand the ML lifecycle before building the solution.
Select and order the steps from the following list to correctly describe the ML lifecycle. Select each step one time.
Show Answer
Correct Answer: Step 1: Prepare the data for training.
Step 2: Train the model.
Step 3: Test the model.
Step 4: Deploy the model.
Explanation: The standard ML lifecycle after data is available is to prepare the data, train the model, evaluate/test its performance, and then deploy the validated model.
Question 98
HOTSPOT
-
A company wants to use ML to increase customer engagement and sales. The company has collected a large dataset that includes customer demographics, purchase history, browsing patterns, and product ratings.
Select the correct ML approach from the following list for each use case. Select each ML approach one time.
Show Answer
Correct Answer: Predict customer lifetime value (CLV) → Regression
Identify likelihood of customer churn → Classification
Group customers by similar purchasing patterns → Clustering
Explanation: Regression predicts continuous numeric values, classification predicts categorical outcomes such as churn/non-churn, and clustering discovers natural customer segments without labels.
Question 99
An AI practitioner is developing a prompt for large language models (LLMs) in Amazon Bedrock. The AI practitioner must ensure that the prompt works across all Amazon Bedrock LLMs.
Which characteristic can differ across the LLMs?
A. Maximum token count
B. On-demand inference parameter support
C. The ability to control model output randomness
D. Compatibility with Amazon Bedrock Guardrails
Show Answer
Correct Answer: A
Explanation: Different Amazon Bedrock foundation models can have different context window sizes and maximum input/output token limits, so the maximum token count is not consistent across all LLMs. Controls for randomness (such as temperature/top_p) are generally available across supported models, Guardrails are a Bedrock feature rather than a model-specific capability, and on-demand inference is a deployment option rather than the key prompt characteristic that varies across models.
Question 100
A company must comply with regulatory standards to develop and use trustworthy AI management solutions.
Which approach will meet this requirement?
A. Optimize model inference time by using high-powered GPUs for faster processing.
B. Ensure that each AI solution is developed only by technical experts. Do not involve other stakeholders.
C. Constrain transparency and user access to each model’s decision-making process.
D. Ensure fairness, transparency, accountability, and security throughout the lifecycle of each AI solution.
Show Answer
Correct Answer: D
Explanation: Trustworthy AI governance and regulatory compliance emphasize fairness, transparency, accountability, and security across the entire AI lifecycle. The other options either focus on performance rather than governance or contradict core responsible AI principles by limiting stakeholder involvement or transparency.
Question 101
Which statement accurately describes Retrieval Augmented Generation (RAG)?
A. A process that uses large amounts of new data to train large language models (LLMs) to improve LLM performance
B. A process by which large language models (LLMs) reference external authoritative knowledge bases to enhance the relevance and accuracy of LLM responses without re-training
C. A process that limits large language models (LLMs) exclusively to their original training data to improve response speed for business applications without re-training
D. A process that focuses on language translation tasks for businesses that operate in multiple countries
Show Answer
Correct Answer: B
Explanation: Retrieval Augmented Generation (RAG) retrieves relevant information from external knowledge sources and provides it to the LLM as context at inference time, improving the relevance and factual accuracy of responses without retraining the model.
Question 102
A company wants to generate synthetic data responses for multiple prompts from a large volume of data. The company wants to use an API method to generate the responses. The company does not need to generate the responses immediately.
Which solution meets these requirements with the LEAST development effort?
A. Input the prompts into the model. Generate responses by using real-time inference.
B. Use Amazon Bedrock batch inference. Generate responses asynchronously.
C. Use Amazon Bedrock agents. Build an agent system to process the prompts recursively.
D. Use AWS Lambda functions to automate the task. Submit one prompt after another and store each response.
Show Answer
Correct Answer: B
Explanation: Amazon Bedrock batch inference is designed for processing large volumes of prompts asynchronously. Because the responses do not need to be generated immediately, batch inference provides the least development effort compared to building orchestration with Lambda or agents, and is more appropriate than real-time inference for high-volume synthetic data generation.
Question 103
Which foundation model (FM) in Amazon Bedrock can be fine-tuned for text, image, and video comprehension?
A. Amazon Nova Pro
B. Amazon Titan Multimodal Embeddings G1
C. Amazon Titan Text G1 - Express
D. Amazon Nova Micro
Show Answer
Correct Answer: A
Explanation: Amazon Nova Pro is Amazon Bedrock's advanced multimodal foundation model that supports comprehension across text, images, and video and can be fine-tuned for domain-specific tasks. Titan Multimodal Embeddings G1 is for generating embeddings rather than multimodal content understanding, Titan Text G1 - Express is text-only, and Nova Micro is a text-only model.
$19
Get all 443 questions with detailed answers and explanations
Instant download HTML + PDF delivered the moment payment clears.
Secure Stripe checkout we never see or store your card details.
7-day refund if files are defective see our refund policy.
