Amazon AWS Certified AI Practitioner AIF-C01 Practice Test 21
Question 200
The initial scenario was: A developer wants an AI-capable integrated coding environment for building application features. The original control is already in place. Now the system later needs a hosted foundation-model inference endpoint. Which ADDITIONAL action BEST addresses the new condition?
- Describe Kiro as an AWS compliance evidence repository
- A developer demonstrates an assisted project workflow and reviewed code changes
- Select the managed model access service for the production integration
- Identify Kiro as an AI-assisted development environment
Explanation
Correct Answer
C:The system later needs a hosted foundation-model inference endpoint. The appropriate added action is: Select the managed model access service for the production integration.
Incorrect Answers
A:This original shortcut remains unsuitable: A coding environment is not a compliance report portal. It is not a valid solution to the changed requirement.
B:This describes evidence for the original approach, not a new action: A developer demonstrates an assisted project workflow and reviewed code changes.
D:This was the original control and is already implemented. Kiro helps developers plan and develop software with AI capabilities.
Question 201
An application team wants an open-source SDK for composing tool-using AI agents in code. Which option BEST meets the stated AI/ML or AWS requirement?
- Evaluate Strands Agents for SDK-based agent construction
- Assess Bedrock AgentCore runtime hosting requirements
- Treat a vector database as the full agent orchestration framework
- Evaluate Amazon Quick for the supported enterprise AI assistant use case
Explanation
Correct Answer
A:Strands is an agent development framework rather than a standalone knowledge database. In this situation, choose: Evaluate Strands Agents for SDK-based agent construction.
Incorrect Answers
B:This is appropriate only after a distinct change: The team then needs operational production hosting rather than SDK constructs. It is not the best initial response to an application team wants an open-source sdk for composing tool-using ai agents in code.
C:Vector storage alone does not coordinate tools and agent behavior. The proposal, Treat a vector database as the full agent orchestration framework, does not satisfy this need.
D:This can be valid for a different task: A business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline. The current request is different and needs the selected approach.
Question 202
An application team wants an open-source SDK for composing tool-using AI agents in code. A colleague suggests: Treat a vector database as the full agent orchestration framework. Which finding BEST explains why the suggestion should be rejected?
- Vector storage alone does not coordinate tools and agent behavior
- The shortcut is safe because its terminology is familiar to the team
- More generated tokens generally change usage charges
- A new foundation model is unnecessary for a governed enterprise search assistant
Explanation
Correct Answer
A:Vector storage alone does not coordinate tools and agent behavior. This is the direct defect in the proposed action: Treat a vector database as the full agent orchestration framework.
Incorrect Answers
B:Familiar wording is not proof that an action is appropriate. The actual defect is: Vector storage alone does not coordinate tools and agent behavior.
C:This concern belongs to another situation: A product owner finds that long responses cost more under a token-metered generative API.
D:This finding is associated with a different problem: A business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline.
Question 203
An application team wants an open-source SDK for composing tool-using AI agents in code. After implementing Evaluate Strands Agents for SDK-based agent construction, which observation BEST verifies the intended result?
- Pilot users receive relevant answers from appropriately permitted information
- A prototype uses the framework to plan authorized tool calls
- The team approves Evaluate Strands Agents for SDK-based agent construction but does not run an operational test
- The team schedules a future test of whether a prototype uses the framework to plan authorized tool calls, but has no result
Explanation
Correct Answer
B:The observed result, A prototype uses the framework to plan authorized tool calls, provides direct evidence. Strands is an agent development framework rather than a standalone knowledge database.
Incorrect Answers
A:This observation would be evidence for another case: A business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline, not the current scenario.
C:Approval is a plan, not a demonstrated result. The required observable condition is: A prototype uses the framework to plan authorized tool calls.
D:An unperformed future check is not evidence. Verify the actual outcome: A prototype uses the framework to plan authorized tool calls.
Question 204
The initial scenario was: An application team wants an open-source SDK for composing tool-using AI agents in code. The original control is already in place. Now the team then needs operational production hosting rather than sdk constructs. Which ADDITIONAL action BEST addresses the new condition?
- Assess Bedrock AgentCore runtime hosting requirements
- Evaluate Strands Agents for SDK-based agent construction
- A prototype uses the framework to plan authorized tool calls
- Treat a vector database as the full agent orchestration framework
Explanation
Correct Answer
A:The team then needs operational production hosting rather than SDK constructs. The appropriate added action is: Assess Bedrock AgentCore runtime hosting requirements.
Incorrect Answers
B:This was the original control and is already implemented. Strands is an agent development framework rather than a standalone knowledge database.
C:This describes evidence for the original approach, not a new action: A prototype uses the framework to plan authorized tool calls.
D:This original shortcut remains unsuitable: Vector storage alone does not coordinate tools and agent behavior.
Question 205
A business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline. Which option BEST meets the stated AI/ML or AWS requirement?
- Require training a new foundational model from scratch to answer internal questions
- Model input and output token volumes in the inference cost estimate
- Evaluate Amazon Quick for the supported enterprise AI assistant use case
- Choose SageMaker AI for managed custom ML development
Explanation
Correct Answer
C:AWS provides enterprise AI experiences to work with organizational information. In this situation, choose: Evaluate Amazon Quick for the supported enterprise AI assistant use case.
Incorrect Answers
A:A new foundation model is unnecessary for a governed enterprise search assistant. The proposal, Require training a new foundational model from scratch to answer internal questions, does not satisfy this need.
B:This can be valid for a different task: A product owner finds that long responses cost more under a token-metered generative API. The current request is different and needs the selected approach.
D:This is appropriate only after a distinct change: Analysts later need a custom predictive ML scoring pipeline. It is not the best initial response to a business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline.
Question 206
A business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline. A colleague suggests: Require training a new foundational model from scratch to answer internal questions. Which finding BEST explains why the suggestion should be rejected?
- A new foundation model is unnecessary for a governed enterprise search assistant
- The shortcut is safe because its terminology is familiar to the team
- More generated tokens generally change usage charges
- Provisioning and pay-as-you-go can have different billing tradeoffs
Explanation
Correct Answer
A:A new foundation model is unnecessary for a governed enterprise search assistant. This is the direct defect in the proposed action: Require training a new foundational model from scratch to answer internal questions.
Incorrect Answers
B:The actual defect is: A new foundation model is unnecessary for a governed enterprise search assistant.
C:This finding is associated with a different problem: A product owner finds that long responses cost more under a token-metered generative API.
D:This concern belongs to another situation: A company needs stable high-throughput model performance for a reserved daily window and is comparing purchase models.
Question 207
A business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline. After implementing Evaluate Amazon Quick for the supported enterprise AI assistant use case, which observation BEST verifies the intended result?
- Pilot users receive relevant answers from appropriately permitted information
- The team approves Evaluate Amazon Quick for the supported enterprise AI assistant use case but does not run an operational test
- The team schedules a future test of whether pilot users receive relevant answers from appropriately permitted information, but has no result
- Billing tests correlate representative requests with input and output tokens
Explanation
Correct Answer
A:The observed result, Pilot users receive relevant answers from appropriately permitted information, provides direct evidence. AWS provides enterprise AI experiences to work with organizational information.
Incorrect Answers
B:The required observable condition is: Pilot users receive relevant answers from appropriately permitted information.
C:Verify the actual outcome: Pilot users receive relevant answers from appropriately permitted information.
D:This observation would be evidence for another case: A product owner finds that long responses cost more under a token-metered generative API, not the current scenario.
Question 208
The initial scenario was: A business analyst wants natural-language search and question answering over governed enterprise information without creating a custom training pipeline. The original control is already in place. Now analysts later need a custom predictive ml scoring pipeline. Which ADDITIONAL action BEST addresses the new condition?
- Require training a new foundational model from scratch to answer internal questions
- Choose SageMaker AI for managed custom ML development
- Evaluate Amazon Quick for the supported enterprise AI assistant use case
- Pilot users receive relevant answers from appropriately permitted information
Explanation
Correct Answer
B:Analysts later need a custom predictive ML scoring pipeline. The appropriate added action is: Choose SageMaker AI for managed custom ML development.
Incorrect Answers
A:This original shortcut remains unsuitable: A new foundation model is unnecessary for a governed enterprise search assistant.
C:This was the original control and is already implemented. AWS provides enterprise AI experiences to work with organizational information.
D:This describes evidence for the original approach, not a new action: Pilot users receive relevant answers from appropriately permitted information.
Question 209
A product owner finds that long responses cost more under a token-metered generative API. Which option BEST meets the stated AI/ML or AWS requirement?
- Compare provisioned throughput with on-demand invocation for expected utilization
- Model input and output token volumes in the inference cost estimate
- Evaluate prompt caching where supported to reduce repeated processing
- Assume generation length has no cost impact under token billing
Explanation
Correct Answer
B:Usage-based pricing can vary with token counts and model choice. In this situation, choose: Model input and output token volumes in the inference cost estimate.
Incorrect Answers
A:This can be valid for a different task: A company needs stable high-throughput model performance for a reserved daily window and is comparing purchase models. The current request is different and needs the selected approach.
C:This is appropriate only after a distinct change: A repeatable system prefix is being resent thousands of times. It is not the best initial response to a product owner finds that long responses cost more under a token-metered generative api.
D:More generated tokens generally change usage charges. The proposal, Assume generation length has no cost impact under token billing, does not satisfy this need.
Question 210
A product owner finds that long responses cost more under a token-metered generative API. A colleague suggests: Assume generation length has no cost impact under token billing. Which finding BEST explains why the suggestion should be rejected?
- Regional feature gaps can invalidate the chosen deployment plan
- More generated tokens generally change usage charges
- Provisioning and pay-as-you-go can have different billing tradeoffs
- The shortcut is safe because its terminology is familiar to the team
Explanation
Correct Answer
B:More generated tokens generally change usage charges. This is the direct defect in the proposed action: Assume generation length has no cost impact under token billing.
Incorrect Answers
A:This concern belongs to another situation: A global consumer app must avoid assuming every Bedrock model is available in every AWS Region.
C:This finding is associated with a different problem: A company needs stable high-throughput model performance for a reserved daily window and is comparing purchase models.
D:Familiar wording is not proof that an action is appropriate. The actual defect is: More generated tokens generally change usage charges.