Get Updated Amazon AIP-C01 Exam Questions

QUESTION 1

A legal research company is developing an AI-powered chat assistant that uses Amazon Bedrock with a knowledge base that contains millions of legal documents. When attorneys search for legal case precedents, the initial retrieval returns many documents. The company wants the chat assistant to generate all responses within 3 seconds.

The foundation model (FM) that the chat assistant uses has a context window limitation that allows effective processing for only 10 documents for each query. Currently, the retrieval configuration fetches 50 documents by using semantic search. Attorneys report that highly relevant precedents frequently appear in positions 15-40 of the results.

The company needs a solution to improve the relevance of query results. The solution must meet the company's latency and context window constraints.

Which solution will meet these requirements?

A

Maintain the 50-document retrieval but increase the FM's temperature parameter. Implement prompt engineering techniques that instruct the model to identify the most relevant documents from the full result set before the model generates responses.

B

Enable reranking on the knowledge base by using a reranker model with language-specific configurations. Configure the chat assistant to initially retrieve 50 documents. Use the reranker model's cross-encoder architecture to evaluate query-document pairs. Return only the top 10 highest-scored documents.

C

Configure three separate knowledge bases. Configure one knowledge base to use 500 token chunks, another to use 1,000 token chunks, and the third to use 2,000 token chunks. Implement a pre-processing action group by using an AWS Lambda function that retrieves 50 documents from each knowledge base, merges the 150 total results, and uses term frequency - inverse document frequency (TF-IDF) scoring to select the top 10 documents.

D

Implement a post-processing action group by using an AWS Lambda function that retrieves 50 documents and calculates cosine similarity scores between the query embedding and each document embedding by using Amazon Titan Embeddings. Configure the action group to sort by similarity score and return the top 10 documents.

Correct Option: B

✅ Option B (Correct) Reasoning: Reranking significantly improves retrieval relevance by re-evaluating the initially retrieved 50 documents with a more sophisticated cross-encoder model. This process ensures the top 10 most relevant documents are passed to the Foundation Model (FM), respecting its context window limitation and improving overall response quality and latency. ❌ Why the other choices are incorrect:

Option A is incorrect: The FM's context window limitation of 10 documents means feeding 50 documents will cause issues. Prompt engineering cannot overcome this hard constraint, and it will likely increase latency or lead to ineffective processing.

Option C is incorrect: Managing multiple knowledge bases is overly complex. TF-IDF is a lexical search method, less effective than semantic search for relevance, and will likely add significant latency, failing the 3-second requirement.

Option D is incorrect: A post-processing action group operates after the FM generates its response. The requirement is to improve the input to the FM. Recalculating cosine similarity on already embedded documents is redundant; a dedicated reranker is more effective and efficient for re-ranking.

Reference: https://docs.aws.amazon.com/bedrock/latest/userguide/knowledge-base-reranking.html

QUESTION 2

A company is implementing a 2-week initiative to incorporate GenAI into several hundred existing applications to provide context-driven natural language help for users. To simplify the process to choose an appropriate model for each application, a GenAI Developer is building a system to automatically compare foundation models (FMs) based on general text generation accuracy. The system runs a job that performs a series of steps.

First, the system creates a service role and attaches an IAM permissions policy that allows Amazon Bedrock to invoke the FMs and grants access to Amazon S3. Then the system requests and receives access to all the FMs that need to be tested. Then the system creates S3 buckets to store the evaluation results. Then the system creates evaluation configurations by using the TREX built-in dataset. Finally, the system specifies the Accuracy metric and Generation evaluation type in each evaluation configuration.

When the GenAI developer attempts to run the system, it fails with an AccessDenied exception. The GenAI developer must resolve the issue.

Which solution will meet this requirement?

A

Create a trust policy that defines Amazon Bedrock as the service principal. Attach the trust policy to the service role.

B

Configure cross-origin resource sharing (CORS) permissions that allow GET and DELETE actions on the S3 buckets.

C

Change the test dataset from TREX to Gigaword in the evaluation configurations.

D

Change the evaluation type to Summarization and the metric type to Correctness in the evaluation configurations.

Correct Option: A

✅ Option A (Correct) Reasoning: For Amazon Bedrock to assume an IAM service role and utilize its attached permissions, a trust policy must be defined on that role. This trust policy explicitly grants bedrock.amazonaws.com permission to assume the role. Without it, Bedrock cannot assume the role, leading to an AccessDenied exception despite correct permissions being attached to the role itself.❌ Why the other choices are incorrect:

Option B is incorrect: CORS permissions are for browser-based cross-origin requests, not for an AWS service like Amazon Bedrock assuming a role or accessing S3 directly for evaluation.

Option C is incorrect: Changing the dataset will not resolve an AccessDenied exception, as it's an issue with permissions or role assumption, not the data source itself.

Option D is incorrect: Modifying the evaluation type or metric type addresses the evaluation methodology, not the underlying AccessDenied error caused by insufficient IAM role trust. The original settings (Generation, Accuracy) are valid for text generation accuracy.

Reference: https://docs.aws.amazon.com/bedrock/latest/userguide/prereqs-iam.html#prereqs-iam-permissions

QUESTION 3

A social messaging company is building an Al chat assistant by using Amazon Bedrock. The company must ensure that every inference complies with an approved safety policy. The company wants to block harmful prompts before model invocations, filter streamed model outputs in real time, and route flagged cases for human review.

Which solution will meet these requirements?

A

Configure an AWS Step Functions workflowe Configure a step to use an AWS Lambda function to pre-check inputs by using the ApplyGuardrail API. Use an InvokeModelWithResponseStream API step that has the guardrail attached. Configure a second Lambda function step to post-check outputs by using the ApplyGuardrail API. Route flagged items to an Amazon SQS queue. Enforce guardrail use by using a bedrock:Guardrailldentifier IAM condition.

B

Configure an AWS Step Functions workflow. Configure a step to call the ApplyGuardrail API before inference. Then call the InvokeModel API without streaming and guardrail. Store the results in Amazon S3. Use the client IJI to hide problematic tokens.

C

Configure an AWS Step Functions workf10VL Configure a step to use the InvokeModelWithResponseStream API that has the guardrail attached for in-stream filtering. Run an AWS Lambda post-check step by using the ApplyGuardrail API to check flagged cases. Do not perform pre-inference.

D

Configure an AWS Step Functions workflow that includes steps to perform pre-checks and post-checks by using the ApplyGuardrail API and the InvokeModelWithResponseStream API. Attach the guardrail to the check steps. Use process controls such as code reviews instead of IAM enforcement to ensure that guardrails are always applied.

Correct Option: A

The company requires blocking harmful prompts before model invocations, filtering streamed model outputs in real time, and routing flagged cases for human review. Option A addresses all these requirements effectively. It uses an AWS Lambda function with the ApplyGuardrail API for pre-inference input checks. It then leverages the InvokeModelWithResponseStream API with the guardrail directly attached, ensuring real-time filtering of model outputs. A second Lambda function can then process the results, specifically routing flagged items (as identified by the guardrail) to an Amazon SQS queue for human review. Crucially, it enforces guardrail use with an IAM condition (bedrock:Guardrailldentifier), which is a robust security control.

Reference: https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails.html

QUESTION 4

A healthcare company is developing a clinical decision support system. The system must use generative Al (GenAI) to analyze patient information and produce recommendations. The system must integrate with existing electronic record systems by using standardized healthcare APIs.

The system must perform real-time detection of sensitive personal health information (PHI) across structured and unstructured content. The system must maintain detailed audit trails for regulatory requirements. The system must be able to scale to 5,000 concurrent requests with minimal infrastructure management.

Which solution will meet these requirements with the LEAST operational overhead?

A

Use Amazon Comprehend Medical to detect PHI. Configure custom AWS Lambda functions to redact sensitive data. Store clinical data in Amazon RDS by using custom schemas. Use Amazon OpenSearch Service to maintain audit trails.

B

Deploy a fully managed Fast Healthcare Interoperability Resources (FHIR) server from AWS Marketplace on Amazon ECS on AWS Fargate. Integrate with Amazon Bedrock Guardrails for PHI detection. Use AWS CloudTrail Lake to store audit logs.

C

Run custom Fast Healthcare Interoperability Resources (FHIR) servers on Amazon ECZ Store clinical data in Amazon S3. Use Amazon Macie to scan S3 buckets for PHI. Capture all access logs by using AWS CloudTrail.

D

Use a fully managed prebuilt healthcare API server from AWS Marketplace that runs on Amazon API Gateway. Use AWS Lambda for API operations. Integrate Amazon Bedrock Guardrails for real-time PHI detection and redaction. Store compliance events and violations in Amazon DynamoDB. use AWS Glue with Amazon Athena to generate audit and regulatory reports.

Correct Option: D

A healthcare company needs a generative AI system with real-time PHI detection, standardized healthcare API integration, detailed audit trails, and scalability for 5,000 concurrent requests with the LEAST operational overhead.

Option D provides a comprehensive serverless solution:

Integration & Scalability: A fully managed prebuilt healthcare API server from AWS Marketplace running on Amazon API Gateway with AWS Lambda for API operations offers high scalability (5,000 concurrent requests) and minimal infrastructure management due to its serverless nature. This approach inherently supports standardized healthcare APIs through the Marketplace offering.
PHI Detection: Amazon Bedrock Guardrails provides real-time PHI detection and redaction, directly integrating with generative AI workflows, which is crucial for the system's purpose.
Audit Trails: Storing compliance events and violations in Amazon DynamoDB (a fully managed, highly scalable NoSQL database) and using AWS Glue with Amazon Athena for audit reports (both serverless analytics services) ensures detailed audit trails with low operational overhead.

This combination of services maximizes serverless components, directly addressing the requirement for the least operational overhead.

Reference: https://aws.amazon.com/api-gateway/, https://aws.amazon.com/lambda/, https://aws.amazon.com/bedrock/guardrails/, https://aws.amazon.com/dynamodb/, https://aws.amazon.com/glue/, https://aws.amazon.com/athena/

QUESTION 5

A retail company is developing 8 conversational Al assistant that uses Amazon Bedrock foundation models (FMs). The Al assistant must handle customer product inquiries. Customers often ask about multiple products in the same conversation-

The Al assistant must maintain context about which product a customer is discussing throughout a conversation. The solution must support high concurrent user volumes with low response times. Data privacy regulations require the company to encrypt all stored customer product preferences at rest. The solution must minimize operational costs and scale automatically during peak shopping periods.

Which solution will meet these requirements?

A

Use Amazon Comprehend built-in entity recognition capabilities to extract product mentions from conversations. Store the entities in Amazon DynamoDB. Set a TTL for the entities. Include the most recent product entity as conversational context in each prompt to the Amazon Bedrock FMs.

B

Create an Amazon Comprehend custom entity recognizer that is trained specifically on product taxonomies. Build a session management service that uses Amazon DynamoDB to provide state persistence. Configure server-side encryption and automatic scaling. Include session context data in prompts to the Amazon Bedrock FMs-

C

Use Amazon Lex to create intents for each product category with predefined slot types. Configure a fallback AWS Lambda function to handle transitions between product discussions. Pass conversation history directly to Amazon Bedrock to track conversational context.

D

Manage memory directly in prompts by including the entire conversation history in each request to the Amazon Bedrock FMs. Apply token prioritization techniques to emphasize recent product mentions. Use Amazon CloudWatch to monitor token usage costs.

Correct Option: B

✅ Option B (Correct)
This solution effectively addresses all requirements. An Amazon Comprehend custom entity recognizer, trained on product taxonomies, accurately extracts specific product mentions, crucial for maintaining context in multi-product inquiries. Using Amazon DynamoDB for session management provides low-latency, highly scalable, and cost-effective state persistence for high concurrent user volumes. DynamoDB inherently supports server-side encryption at rest, meeting data privacy regulations, and offers automatic scaling during peak periods. Including this rich session context in prompts to Amazon Bedrock FMs ensures the AI assistant maintains conversational flow efficiently and accurately.

❌ Why the other choices are incorrect:

Option A is incorrect: Amazon Comprehend's built-in entity recognition might lack the specificity needed for unique product taxonomies, leading to less accurate context. While DynamoDB with TTL is good for state, a custom recognizer offers superior accuracy for the domain.
Option C is incorrect: Using Amazon Lex to create intents for each product category adds significant complexity if the core conversational logic is handled by Bedrock FMs. Passing the entire conversation history directly to Bedrock for context can become very expensive and hit token limits quickly, failing to minimize operational costs.
Option D is incorrect: Managing memory by including the entire conversation history in each prompt is inefficient and costly. It drastically increases token usage, making it expensive and susceptible to context window limitations, directly contradicting the requirement to minimize operational costs.

Reference: https://docs.aws.amazon.com/comprehend/latest/dg/comprehend-cer.html

QUESTION 6

A financial services company uses Amazon Bedrock to analyze customer data that is stored in an Amazon S3 bucket. The data includes personally identifiable information (PII). The company must mask PII from foundation model (FM) responses.

Which solution will meet this requirement with the LEAST operational effort?

A

Create a guardrail in Amazon Bedrock to filter PII content. Define the PII type and set the guardrail action to MASK. Configure Amazon Bedrock to apply the filter to each FM response.

B

Use Amazon Comprehend to detect PII entities in the S3 data before invoking Amazon Bedrock. Configure an AWS Lambda function to call the Amazon Comprehend DetectPiiEntities API to mask detected PII. Store the processed data back to the original S3 bucket.

C

Use Amazon Macie to scan the S3 bucket for PII data. Configure an AWS Lambda function to store PII in a second S3 bucket. Use an Amazon EventBridge rule to invoke the Lambda function.

D

Configure an AWS Lambda function to search for PII data. Implement a step in the Lambda function code to store the PII in a second S3 bucket and non-PII data into a third S3 bucket.

Correct Option: A

Option A is the correct solution because Amazon Bedrock Guardrails are specifically designed to filter and mask sensitive content, including Personally Identifiable Information (PII), from both user inputs and foundation model (FM) responses. By creating a guardrail, defining PII types, and setting the action to MASK, Bedrock handles the PII masking as part of its managed service, requiring the least operational effort compared to building custom solutions or integrating multiple external services.

Reference: https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails.html

QUESTION 7

A cooking spice company is launching a marketing campaign that will invite customers to upload videos of themselves cooking with the company's spices to a platform that the company operates. The company is designing a solution for the video sharing platform.

The solution must extract video transcripts and generate embeddings to help users search for videos based on spices, other ingredients, and cooking methods. The solution must integrate with the company's existing open source chat system so users can interact with videos in real time. The solution must analyze user chat text for sentiment to determine the most positively received videos. Each month, the recipe videos that received the most positive feedback and the highest number of views will be displayed on the platform.

Which combination of steps will meet these requirements with the LEAST operational overhead? (Select THREE.)

A

Use Amazon Transcribe to perform automatic speech recognition (ASR) to extract video transcripts. Use Amazon Comprehend to perform sentiment analysis on chat feedback. Use an Amazon Bedrock embeddings model to generate embeddings. Store the embeddings in an Amazon Bedrock knowledge base to enable semantic retrieval.

B

Use AWS Step Functions to orchestrate the entire application pipeline, Configure a step in the workflow to use an AWS Lambda function to call the chat system API.

C

Configure a custom orchestration AWS Lambda function to call all application components directly, manage errors, and embed custom code in calls to the chat system API,

D

Store video metrics such as positive feedback counts and view counts in Amazon DynamoDB,

E

Use the Strands Agents SDK with built-in HTTP tools to call the open source chat system API. Use the Strands Agents SDK built-in orchestration capabilities to coordinate calls to Amazon Bedrock Knowledge Bases to enable video search and to Amazon Comprehend to perform sentiment analysis.

F

Store video metrics such as positive feedback counts and view counts in an Amazon Aurora PostgreSQL database that uses the pgvector extension.

Correct Option: A,B,D

Option A (Correct): Amazon Transcribe extracts video transcripts with minimal overhead. Amazon Comprehend performs sentiment analysis on chat feedback. Amazon Bedrock embeddings models and knowledge bases provide fully managed services for generating embeddings and enabling semantic search, significantly reducing operational burden.
Option B (Correct): AWS Step Functions provides a serverless workflow orchestrator for the entire application pipeline, managing state and error handling. An AWS Lambda function is a cost-effective, serverless way to call the existing chat system API, ensuring low operational overhead.
Option D (Correct): Amazon DynamoDB is a fully managed, serverless NoSQL database ideal for storing high-throughput, low-latency data like video metrics (feedback counts, view counts). Its automatic scaling and minimal administration make it an excellent choice for least operational overhead.
Option C (Incorrect): A custom orchestration Lambda function would require significant custom code for error handling, state management, and retries, leading to higher operational overhead compared to AWS Step Functions.
Option E (Incorrect): "Strands Agents SDK" is not a recognized AWS service. Relying on a hypothetical or third-party SDK for core functionality would likely increase operational overhead and complexity, contradicting the requirement for an AWS-centric solution with least overhead.
Option F (Incorrect): While Amazon Aurora PostgreSQL with pgvector can store embeddings, using it for simple feedback and view counts is higher overhead than DynamoDB. DynamoDB is more suitable and cost-effective for simple key-value metrics storage with minimal operational effort.

Reference: https://aws.amazon.com/transcribe/, https://aws.amazon.com/comprehend/, https://aws.amazon.com/bedrock/, https://aws.amazon.com/step-functions/, https://aws.amazon.com/lambda/, https://aws.amazon.com/dynamodb/

QUESTION 8

A company uses an Anthropic Claude Haiku model in Amazon Bedrock to power an application that answers customer questions about the company's products. Approximately 60% of the questions that the application receives each day are very similar.

The company notices that the application cost exceeds the monthly budget by nearly 50%. Amazon CloudWatch metrics show high input and output token counts. The company needs a solution to reduce the application cost.

Which solution will meet this requirement with the LEAST deployment effort?

A

Use the prompt caching feature in Amazon Bedrock and define cache checkpoints for application prompts. Define the minimum and maximum numbers of cache checkpoints.

B

Enable simplified cache management for Claude models in Amazon Bedrock. Use multiple cache checkpoints to provide granular control.

C

Use an Amazon ElastiCache cluster to store the repeated questions. Configure an AWS Lambda function to check ElastiCache for cache hits and to call Amazon Bedrock for cache misses.

D

Configure an AWS Lambda function to use custom code to store repeated questions in an Amazon DynamoDB table. Enable vector search for the table. Create a second Lambda function to check DynamoDB first for cache hits before invoking Amazon Bedrock.

Correct Option: A

The core problem is high application cost due to high input/output token counts from approximately 60% similar customer questions. The company needs a solution with the LEAST deployment effort. Amazon Bedrock's native prompt caching feature is designed precisely for this scenario.

✅ Option A is correct: Amazon Bedrock offers a prompt caching feature that allows the service to cache common prompt prefixes and reuse previously generated responses up to certain 'cache checkpoints.' This significantly reduces the number of tokens sent to the model and generated, directly addressing the cost issue. Configuring this feature within Bedrock involves minimal deployment effort compared to building custom caching solutions. The concept of 'defining cache checkpoints' and controlling their 'minimum and maximum numbers' aligns with how users can manage the granularity of caching (e.g., by specifying stop sequences or token lengths) to optimize cost and latency.

❌ Option B is incorrect: While this option also refers to Bedrock's prompt caching, stating 'simplified cache management' and 'multiple cache checkpoints to provide granular control' is generally true, Option A provides a slightly more specific and actionable description of the configuration aspects ('define cache checkpoints' and 'minimum and maximum numbers') that align with Bedrock's caching mechanisms for optimizing efficiency.
❌ Option C is incorrect: Implementing a custom caching solution with Amazon ElastiCache and AWS Lambda functions involves significant deployment and operational effort, including setting up and managing an ElastiCache cluster, writing and deploying Lambda code, and integrating it into the application logic. This contradicts the requirement for the 'LEAST deployment effort.'
❌ Option D is incorrect: A custom solution using Amazon DynamoDB with vector search and two AWS Lambda functions is even more complex than option C. It requires substantial development, infrastructure setup (DynamoDB table, vector index), and management, making it the highest deployment effort among the choices.

Reference: https://aws.amazon.com/blogs/machine-learning/reduce-latency-and-cost-for-generative-ai-applications-with-prompt-caching-on-amazon-bedrock/

QUESTION 9

A media company is building an AI-powered content moderation system by using Amazon Bedrock. The system first classifies text by using a small, low-latency model. Then the system escalates requests that have a confidence score below 0.65 to a larger, more expensive model.

The system must respond in near real time for high-confidence results. The system must process low-confidence requests asynchronously. The system must scale to meet sudden spikes in demand. The company wants to optimize costs for the system by invoking the larger model only when required. The company wants to use decoupled components to achieve high resiliency for the system.

Which solution will meet these requirements?

A

Use Amazon API Gateway to invoke the small model synchronously. If the small model's confidence score is below 0.65, synchronously call the larger model. Use provisioned concurrency to handle traffic spikes.

B

Use an AWS Step Functions workflow that has parallel branches to run both the small model and the large model for every request. Choose the large model result when confidence score values differ.

C

Send requests to an Amazon SQS queue. Use AWS Fargate to process messages. Invoke the small model first. If the confidence score is below 0.65, place the request in a second SQS queue to process asynchronously by using the large model.

D

Deploy both models on Amazon EC2 instances and enable auto scaling. Use a custom application heuristic to route requests to the appropriate instance based on phrase length and keyword rules.

Correct Option: C

The solution must classify text using a small, low-latency model first, and only escalate to a larger, more expensive model if the confidence score is below 0.65. High-confidence results need near real-time response, while low-confidence requests must be processed asynchronously. The system needs to scale for spikes, optimize costs, and use decoupled components for resiliency.

Option C best meets these requirements:

Sending requests to an Amazon SQS queue and processing them with AWS Fargate provides high scalability and decoupling, handling sudden spikes.
Invoking the small model first with Fargate, and if the confidence is high, the result can be provided. This path, while using SQS, can be optimized for near real-time processing through efficient queue polling and Fargate scaling.
If the confidence score is below 0.65, placing the request in a second SQS queue for asynchronous processing by the large model perfectly meets the asynchronous processing, cost optimization (large model only invoked when necessary), and decoupling requirements.

Reference: https://aws.amazon.com/builders-library/designing-event-driven-systems/ https://docs.aws.amazon.com/AWSSimpleQueueService/latest/SQSDeveloperGuide/welcome.html https://docs.aws.amazon.com/AmazonECS/latest/developerguide/fargate.html

QUESTION 10

A company is using Amazon Bedrock to build an AI assistant to help internal teams analyze unstructured customer feedback data. The company stores the customer feedback in an Amazon S3 bucket. The S3 bucket contains more than 25 TB of historical data from mobile app reviews, chat conversations, and call center transcripts. The company expects the data source to grow by 3 GB every day. The data entries often contain multiple unrelated topics within the same input.

The company needs a solution that reliably delivers accurate answers to questions based on the data source. The solution must not export any personally identifiable information (PII) to the Amazon Bedrock model during processing or response generation.

Which solution will meet these requirements with the LEAST operational overhead?

A

Configure an AWS Lambda function that processes each new file in the S3 bucket to detect and remove PII by using Amazon Comprehend. Configure the function to generate fixed-size chunk embeddings and store them in an Amazon OpenSearch Serverless vector store. Configure a second Lambda function to process questions, retrieve context, and invoke an Amazon Bedrock foundation model directly to generate answers.

B

Configure an Amazon Bedrock knowledge base that synchronizes with the S3 bucket by using fixed-size chunking. Configure the knowledge base to use an Amazon Aurora PostgreSQL vector store. Configure an Amazon Bedrock Guardrail to block all types of PII during input and output processing. Configure Amazon Bedrock AgentCore to use the knowledge base and the guardrail to process and answer queries.

C

Configure an Amazon Bedrock knowledge base that synchronizes with the S3 bucket by using semantic chunking. Configure the knowledge base to use an Amazon OpenSearch Serverless vector store. Configure an Amazon Bedrock Guardrail to block all types of PII during input and output processing. Configure Amazon Bedrock AgentCore to use the knowledge base and the guardrail to process and answer queries.

D

Configure an Amazon Bedrock knowledge base that synchronizes with the S3 bucket by using semantic chunking. Configure the knowledge base to use an Amazon OpenSearch Serverless vector store. Configure an Amazon Bedrock Guardrail to block all types of PII during output processing only. Configure Amazon Bedrock AgentCore to use the knowledge base and the guardrail to process and answer queries.

Correct Option: C

Option C presents the solution with the LEAST operational overhead while meeting all requirements. Let's break down why:

Amazon Bedrock Knowledge Base and AgentCore: These are fully managed services designed for building RAG-based AI assistants. They significantly reduce operational overhead compared to custom solutions involving AWS Lambda functions for data processing, chunking, embedding generation, context retrieval, and model invocation.
Semantic Chunking: The problem states that data entries often contain multiple unrelated topics within the same input. Semantic chunking is crucial here because it intelligently groups semantically related sentences or paragraphs, even if they are not fixed in size. This ensures that the retrieved context is coherent and relevant, leading to more accurate answers. Fixed-size chunking (Option B) is less effective for this data type.
Amazon OpenSearch Serverless: This is a fully managed, scalable, and serverless vector store solution, perfectly suited for large and growing datasets like 25+ TB with 3 GB/day growth. It offers low operational overhead compared to managing Aurora PostgreSQL (Option B) for vector search at this scale.
Amazon Bedrock Guardrail (Input and Output): The requirement explicitly states that PII must NOT be exported to the Amazon Bedrock model during processing OR response generation. A Guardrail configured to block PII during both input and output processing ensures that PII is filtered before it reaches the foundation model as part of the prompt (including retrieved context) and also before the final response is delivered to the user. This is a managed and effective way to handle PII without custom code.

Why other options are incorrect:

Option A is incorrect: This approach uses custom AWS Lambda functions for almost every step (PII detection, chunking, embedding, context retrieval, model invocation). This creates significant operational overhead due to the need for custom code development, deployment, and maintenance, which directly contradicts the

Reference: https://aws.amazon.com/bedrock/knowledge-bases/, https://aws.amazon.com/bedrock/agents/, https://aws.amazon.com/bedrock/guardrails/, https://docs.aws.amazon.com/bedrock/latest/userguide/kb-chunking.html, https://docs.aws.amazon.com/opensearch-service/latest/developerguide/serverless.html

QUESTION 11

A company is designing a GenAI solution to help children between the ages of 8 and 14 understand world history. The system must support concurrent use for up to 30 children in a classroom and provide responses in less than 1 second. The company is using Amazon Bedrock and Amazon Nova Pro with a knowledge base that contains history textbooks that are written for students between the ages of 6 and 21. The company uses an Amazon Bedrock flow to ensure that the solution uses the knowledge base to generate every response. The company sets the prompt response temperature to 0.7.

Early testing of the solution results in some responses that are historically inaccurate, violent, and not appropriate for children. A GenAI developer must prevent the solution from responding to prompts inaccurately and inappropriately.

Which solution will meet this requirement?

A

Increase the inference temperature in the flow. Add a node to the flow after the initial response that calls the Amazon Nova model to remove any violence from the response. Return the result of the model call.

B

Configure Amazon Bedrock guardrails for the model responses by using content filters. Set the content filter strength high and set the guardrail action to block. Add the guardrail to the flow. Reduce the inference temperature in the flow.

C

Use Amazon Bedrock Data Automation with a custom output that removes all violent content from the knowledge base. Increase the inference temperature in the flow.

D

Configure Amazon Bedrock guardrails for the prompts by using content filters. Set the content filter strength to high and set the guardrail action to block. Add the guardrail to the flow. Reduce the inference temperature in the flow.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 12

A company is developing a customer support chat assistant that uses an Amazon Bedrock foundation model (FM). The company wants to update to a newer FM version but needs to implement a validation system to detect semantic drift in responses. The company wants to ensure that performance and functionality for end users remains consistent.

The company needs a solution to compare responses between current and new FM versions for 500 test cases. The solution must detect changes in response meaning, generate quantitative similarity scores, complete validations, and log detailed results for historical comparison.

Which solution will meet these requirements with the LEAST operational complexity?

A

Configure Amazon CloudWatch Synthetics canaries that use custom JavaScript code to send identical prompts to both new and existing FM versions. Store responses in Amazon S3. Perform manual reviews to identify semantic differences between FM versions.

B

Configure an AWS Step Functions workflow that sends test prompts to both new and existing FM versions. Use Amazon Bedrock embedding models to calculate cosine similarity scores. Store the results in Amazon DynamoDB with a composite key schema.

C

Use Amazon Bedrock model evaluation jobs to compare the new FM version against the current version using the 500 test cases. Configure the evaluation to calculate semantic similarity metrics. Store results in Amazon S3 for historical comparison.

D

Build a custom solution by using Amazon SageMaker AI to train a classifier model on historical responses to detect anomalies when the solution compares responses from the new FM version to previous patterns.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 13

A company needs a system to automatically generate study materials from multiple content sources. The content sources include document files (PDF files, PowerPoint presentations, and Word documents) and multimedia files (recorded videos). The system must process more than 10,000 content sources daily with peak loads of 500 concurrent uploads. The system must also extract key concepts from document files and multimedia files and create and store contextually accurate summaries. The generated study materials must support real-time collaboration with version control. Which solution will meet these requirements?

A

Use Amazon Bedrock Data Automation (BDA) with AWS Lambda functions to orchestrate document file processing. Use Amazon Bedrock Knowledge Bases to process all multimedia. Store the content in Amazon DocumentDB with replication. Collaborate by using Amazon SNS topic subscriptions. Track changes by using Amazon Bedrock Agents.

B

Use Amazon Bedrock Data Automation (BDA) with foundation models (FMs) to process document files. Integrate BDA with Amazon Textract for PDF extraction and with Amazon Transcribe for multimedia files. Store the processed content in Amazon S3 with versioning enabled. Store the metadata in Amazon DynamoDB. Collaborate in real time by using AWS AppSync GraphQL subscriptions with DynamoDB.

C

Use Amazon Bedrock Data Automation (BDA) with Amazon SageMaker AI endpoints to host content extraction and summarization models. Use Amazon Bedrock Guardrails to extract content from all file types. Store document files in Amazon Neptune for time series analysis. Collaborate by using Amazon Bedrock Chat for real-time messaging.

D

Use Amazon Bedrock Data Automation (BDA) with AWS Lambda functions to process batches of content files. Fine-tune foundation models (FMs) in Amazon Bedrock to classify documents across all content types. Store the processed data in Amazon ElastiCache (Redis OSS) by using Cluster Mode with sharding. Use Amazon Bedrock Prompt Management for version control.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 14

A GenAI developer is developing an AI agent by using the open source Strands Agents framework on AWS. The foundation models (FMs) are based on OpenAI models. After successful testing and validation, the GenAI developer wants to deploy the AI agent to the us-east-1 Region to be close to most users.

The AI agent must provide enterprise-grade security and scalability. The AI agent must also provide real-time observability and support processes that run for more than 1 hour. The GenAI developer must integrate the AI agent with the Okta identity provider (IdP). The AI agent must have memory and context awareness to handle complex interactions and learning.

Which solution will meet these requirements in the MOST operationally efficient way?

A

Create an AWS Lambda function. Create an Amazon Bedrock AgentCore deployment and alias. Build a custom workflow to integrate the agent with the Okta IdP and to provide context awareness. Use Amazon CloudWatch Logs to provide monitoring and observability capabilities.

B

Use the open source CrewAI agentic framework to deploy the agent in Amazon Bedrock. Use CrewAI flows to orchestrate agent activities and manage states. Use native AWS integrations to provide observability and authentication.

C

Deploy a containerized AI application on Amazon ECS with the AWS Fargate launch type by using Amazon Bedrock Anthropic FMs. Use Amazon Bedrock AgentCore to provide observability and IdP integration.

D

Instrument and deploy the agent in Amazon Bedrock AgentCore. Use AgentCore Memory to manage context, and use built-in functionalities to integrate with Okta and to provide observability.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 15

A company is developing a generative Al (GenAl) application by using Amazon new data points daily across AWS Regions in Europe, North America, and Asia before storing the data in Amazon S3. The application must comply with local data protection and storage regulations. Data residency and processing must occur within the same continent. The application must also maintain audit trails of the application's decision-making processes and provide data classification capabilities.

Which solution will meet these requirements?

A

Deploy the application in each Region with local IAM policies. Use Amazon Bedrock cross-Region inference to distribute the workload. Use Amazon CloudWatch to log Al decision-making processes and data processing activities. Manually track compliance certifications across Regions.

B

Use SCPs with AWS Organizations to manage location-specific permissions. Use AWS CloudTrail immutable logs to audit the decision-making processes. Import a custom model into Amazon Bedrock and deploy the model to each Region.

C

Use Amazon S3 Object Lock with Region-specific S3 bucket policies. Pre-process the data points within the Region based on geographic origin before sending the data points to Amazon Bedrock. Use Amazon Macie to classify the data. Use AWS CloudTrail immutable logs to audit the decision-making processes.

D

Create separate AWS accounts for each Region with individual compliance frameworks- Use Amazon SageMaker Al with custom monitoring to track model performance and compliance with data residency requirements. Create manual reports for each regulatory jurisdiction.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 16

A company needs a system to automatically generate study materials from multiple content source, e-content sources Include document files (PDF files, PowerPoint presentations, and Word documents) and multimedia files (recorded videos). The system must process more than 10,000 content sources daily with peak loads of 500 concurrent uploads. The system must also extract key concepts from document files and multimedia files and create and store contextually accurate summaries. The generated study materials must support real-time collaboration with version control.

Which solution will meet these requirements?

A

Use Amazon Bedrock Data Automation (BDA) with AWS Lambda functions to orchestrate document file processing. Use Amazon Bedrock Knowledge Bases to process all multimedia. Store the content in Amazon DocumentDB with replication. Collaborate by using Amazon SNS topic subscriptions. Track changes by using Amazon Bedrock Agents.

B

Use Amazon Bedrock Data Automation (BDA) with foundation models (FMs) to process document files. Integrate BDA with Amazon Textract for PDF extraction and with Amazon Transcribe for multimedia files. Store the processed content in Amazon S3 with versioning enabled. Store the metadata in Amazon DynamoDB. Collaborate in real time by using AWS AppSync GraphQL subscriptions with DynamoDB.

C

Use Amazon Bedrock Data Automation (BDA) with Amazon SageMaker Al endpoints to host content extraction and summarization models. Use Amazon Bedrock Guardrails to extract content from all file types. Store document files in Amazon Neptune for time series analysis. Collaborate by using Amazon Bedrock Chat for real-time messaging.

D

Use Amazon Bedrock Data Automation (BDA) with AWS Lambda functions to process batches of content files. Fine-tune foundation models (FMs) in Amazon Bedrock to classify documents across all content types. Store the processed data in Amazon ElastiCache (Redis OSS) by using Cluster Mode with sharding. Use Amazon Bedrock Prompt Management for version control.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 17

A global healthcare company is deploying a GenAl application on Amazon Bedrock to produce treatment recommendations. Regulations vary for each country where the company operates. Some countries require the company to retain all model inputs and outputs for 2 years. Other countries require the company to submit data for local audits only.

Medical providers require consistent medical terminology across all locations. However, the treatment recommendations that the model produces must adapt to local patient demographics. The solution must also integrate with existing electronic health record (EHR) systems.

The application must support up to 10,000 healthcare provider queries every day with sub-second response times. The company must be able to review the application before deployments and approve of prompt changes. The application must produce comprehensive logs for prompts, responses, and user context.

Which solution will meet these requirements?

A

Use AWS CloudTrail to log API calls. Create standard prompts in Amazon Bedrock Prompt Management that include variables for patient demographics. Implement IAM policies to ensure that only approves users can access prompts.

B

Use Amazon CloudWatch Logs to collect detailed model invocation logs. Store the logs in Amazon S3. Create parameterized prompts in Amazon Bedrock Prompt Management that include variables for treatment options. Enable prompt versioning and set up an approval workflow.

C

Create AWS Lambda functions to dynamically generate prompts that enforce clinical language requirements. Use Amazon CloudWatch Logs to track model invocations. Use Amazon SQS queues to implement a prompt approval workflow.

D

Store prompt templates in Amazon S3. Use S3 Object Lock to implement version control- Use Amazon EventBridge to track model invocations- Use AWS Config to monitor changes to prompt templates.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 18

A healthcare company is developing an application that processes sensitive patient data and generates treatment summaries by using a foundation model (FM). The application must maintain an audit trail of all prompts and completions. The application must securely handle protected health information (PHI) data throughout the processing lifecycle. The company must track all prompt-completion pairs with original patient data, but the company must redact any PHI from stored records. The company must enforce healthcare-specific data retention policies.

Which solution will meet these requirements?

A

Store prompt-completion pairs in Amazon S3 and enable default server-side encryption. Deploy an AWS Lambda function that scans records by using Amazon Comprehend Medical. Configure the Lambda function to delete PHI data after a required retention period.

B

Use Amazon Bedrock Prompt Management and Amazon Bedrock Flows to detect PHI. Configure Amazon Bedrock Guardrails to use sensitive information filters to mask PHI automatically. Store logs in Amazon S3 and configure appropriate retention settings.

C

Store prompt-completion pairs in Amazon S3. Configure S3 Object Lock in compliance mode. Apply tag-based S3 Lifecycle policies to PHI data. Configure an AWS Lambda function to redact patient information after the application processes the data.

D

Store patient data in an Amazon Bedrock knowledge base- Use Amazon Comprehend Medical to identify PHI. Configure an AWS Lambda function to manage data retention according to healthcare policies.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 19

A hospital is building an Al application to help medical clinicians to make treatment decisions, The application uses Amazon Bedrock to analyze patient case histories and suggest diagnoses. The application must maintain sub-500 ms response times to integrate with the hospital's existing real-time clinical workflow- To comply with privacy regulations, the application must log all personally identifiable information (PII) handling decisions for audits. The application must detect and remove PII from responses with at least 99% accuracy.

After initial deployment, clinicians report that diagnostic summaries from the application occasionally include patient names and medical record numbers that were not present in the original case history inputs. An investigation reveals that Amazon Comprehend Medical successfully detects and removes PII from inputs with 95% accuracy, and the application replaces all detected entities with tokens before it sends inputs to Amazon Bedrock- However, the application continues to generate patient-identifying information in approximately 3-5% of outputs.

The company needs a solution to prevent the application from displaying PII in outputs while meeting all other operational requirements.

Which solution will meet these requirements?

A

Configure Amazon Bedrock guardrails with sensitive information filters to detect and block PII in model outputs.

B

Implement a secondary PII detection layer by using regular expressions and custom entity recognition to detect identifiers that Amazon Comprehend Medical misses before sending inputs to Amazon Bedrock.

C

Remove detailed medical context from case histories during pre-processing to prevent the model from generating patient-specific information based on clinical pattern associations.

D

Enable session isolation in Amazon Bedrock API calls. Clear conversation history between requests to prevent patient information from persisting across multiple case analyses.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 20

A healthcare company is building an AI assistant that uses Amazon Bedrock to summarize patient case notes. The AI assistant must process 20,000 case notes daily with peak loads of 100 requests every minute. The AI assistant must maintain sub-second pre-processing latency. The company must ensure that all personally identifiable information (PII) is removed from notes before any text is sent to Amazon Bedrock.

The Al assistant must be able to detect PII in both English and Spanish and prevent the foundation model (FM) from returning medical identifiers in generated summaries. The Al assistant must detect PII and provide audit visibility for all redacted fields,

Which solution will meet these requirements?

A

Use Amazon Comprehend to detect PII in English and Spanish case notes and to apply automatic redaction before the Al assistant invokes Amazon Bedrock. Configure Amazon Bedrock guardrails to block disallowed medical identifiers in model outputs. Store redacted input and output logs in Amazon S3 for auditing.

B

Implement a multi-stage PII detection workflow that uses Amazon Comprehend to detect standard PII types. Implement a fine-tuned Amazon Bedrock model to detect specialty medical identifiers. Combine outputs to create a redacted version of each note and then invoke Amazon Bedrock to summarize the notes- Store audit logs in Amazon S3 to record redaction decisions.

C

Use Amazon SageMaker Clarify to detect PII before invoking the FM. Configure AWS WAF rules to block outbound responses that contain sensitive patient data. Store all processed case notes in Amazon DynamoDB.

D

Use AWS Glue DataBrew to apply a one-time masking transformation to all case notes. Send the masked notes to Amazon Bedrock and implement AWS Lake Formation column-level security to restrict downstream access.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 21

A company wants to replace the FAQ section of its website with an Amazon Bedrock AgentCore agent, The agent has access to more than 700,000 customer inquiries from the company's online support forum. Customers add more than 1 ,500 new inquiries every day. The agent must refresh the data that it references daily'. The agent must not refer to personally identifiable information (PII) that users post online.

Which solution will meet these requirements?

A

Use Amazon S3 vectors to create a knowledge base, Set up a web crawler as a data source. Use Amazon Macie to identify PII and configure an AWS Lambda function to redact identified PII. Configure the agent to use RAG to retrieve data from the knowledge base,

B

Use Amazon OpenSearch Serverless to create a knowledge base- Set up a web crawler as a data source. Implement sensitive information filters for the agent. Configure the agent to use RAG to retrieve data from the knowledge base.

C

Use Amazon Aurora PostgreSQL Serverless to create a knowledge base. Set up a web crawler as a custom data source. Prompt the model to block all PII from agent responses- Configure the agent to use RAG to retrieve data from the knowledge base.

D

Use AWS Step Functions to orchestrate a workflow that uses AWS Lambda functions, an Amazon SOS queue, Amazon DynamoDB, and Amazon Comprehend to crawl through the company's forum. Configure the workflow to incrementally import new data into Amazon S3 and redact any PII. Update the model's prompt daily to use the updated data.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 22

A company uses Amazon Bedrock to deploy an application that generates technical documentation for users across multiple AWS Regions and in multiple languages. Users frequently submit semantically similar questions in different languages, which results in increased inference costs and response latency- The company needs a caching solution that significantly reduces inference costs, provides Iow-latency responses globally, maintains cache freshness with a 5-minute TTL, and minimizes custom cache key generation and application-managed caching logic.

Which solution will meet these requirements?

A

Create a custom caching system that uses AWS Lambda functions to store inference results in an Amazon DynamoDB table. Use Amazon CloudFront to distribute cached responses to global users with a 5-minute TTL.

B

Configure prompt caching in Amazon Bedrock for semantically similar queries across languages. Use Amazon CloudFront and Lambda@Edge functions to handle Regional cache distribution. Set a TTL of 5 minutes for both caching layers.

C

Create Amazon ElastiCache (Redis OSS) clusters in each Region where the application runs to store inference results with custom fingerprinting for multilingual queries- Configure automatic replication between Regional clusters with a 5-minute TTL.

D

Use Amazon DynamoDB Accelerator (DAX) to cache inference results and to automatically manage TTL. use Amazon CloudFront to distribute API responses globally. Use edge functions to handle language-specific transformations.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 23

A healthcare company is implementing a clinical knowledge base application on Amazon Bedrock that provides medical information to doctors. During testing, a quality team discovers that the generative AI (GenAI) model occasionally fabricates medical treatment recommendations that do not originate from the approved clinical guidelines.

The quality team needs to implement a solution to detect the hallucinations before releasing the application. The solution must analyze model responses against verified medical information- The solution must identify semantic inconsistencies when users ask similar questions in different ways.

Which solution will meet these requirements?

A

Deploy a real-time monitoring system that uses Amazon CloudWatch Logs Insights to analyze response patterns and to flag factual inaccuracies based on predefined keywords. Implement AWS Lambda functions to compare responses with known answers from an Amazon DynamoDB table-

B

Create a reference dataset with validated question-answer pairs from clinical guideline. Implement output diffing by using Amazon Bedrock Guardrails for response consistency analysis. Use Amazon Bedrock evaluation to detect factual inaccuracies.

C

Configure Amazon Bedrock Guardrails with custom rules to detect and block potentially hallucinated content by identifying specific patterns in the GenAI responses. Use Amazon SageMaker Feature Store to maintain a repository of verified clinical guidelines-

D

Create an Amazon Bedrock automatic model evaluation job with a custom prompt dataset- Set up anomaly detection alarms to identify factual inaccuracies. Implement Amazon SNS notifications when inaccuracies are detected.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 24

A company used Amazon Bedrock to build a customer support AI assistant, The Al assistant handles approximately 100 requests every second through an Amazon API Gateway API, API Gateway proxies requests to an AWS Lambda function that makes requests to an Amazon Bedrock foundation model (FM) and Amazon DynamoDB tables.

The Al assistant is becoming increasingly expensive because customers often send long conversational histories. Many messages repeat the same information or contain low-value details. Token usage has steadily risen, leading to inconsistent latency and unnecessary costs. The company must reduce input token volume without degrading answer quality and without major changes to the existing architecture.

Which solution will meet these requirements?

A

Replace the current FM with a higher-reasoning model family that supports a larger context window.

B

Configure a new Lambda function to prune repeated or low-value content, summarize long histories by using a small embedding model, and to set maximum input and output token limits in the Amazon Bedrock InvokeModel API request.

C

For each customer interaction, store the complete conversation history in 8 DynamoDB table. Send the full history to the FM with every request.

D

Index all conversations in Amazon Kendra so the model can search across the full conversational context before it generates each response.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 25

A company updates an Amazon Bedrock based assistant on each development sprint, The company needs an automated deployment validation system, The system must perform prerelease validation and control releases automatically. The system must continuously monitor post-deployment. The system must include alarm capabilities and an automatic rollback functionality.

The system must replay synthetic user workflows for key intents- The system must conduct Al-specific evaluations against the production baseline. For example, the evaluations must include hallucination rate and semantic drift analysis. The system must perform automated consistency checks by using a reference prompt set.

The company will use an AWS CodePipeline pre-deploy stage that calls an AWS Step Functions workflow to orchestrate the validation checks.

Which Step Functions workflow configuration will meet these requirements?

A

Run Amazon Bedrock model evaluations and Amazon CloudWatch Synthetics canaries. Run embedding-based consistency checks. Enforce automatic approval or rollback. Schedule nightly runs by using CloudWatch alarms.

B

Run Amazon Bedrock model evaluations and consistency checks- Enforce manual approval. Schedule weekly runs without alarms.

C

Run Amazon CloudWatch Synthetics canaries and consistency checks only. Enforce automatic rollback on failures. Schedule hourly runs without model evaluations.

D

Run Amazon Bedrock model evaluations and Amazon CloudWatch Synthetics canaries. Enforce automatic approval on success. Schedule post-deploy runs with alarms and performance metric tracking.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 26

A healthcare company is building an Al assistant that uses Amazon Bedrock to summarize patient case notes. The Al assistant must process 20,000 case notes daily with peak loads of 100 requests every minute. The Al assistant must maintain sub-second pre-processing latency- The company must ensure that all personally identifiable information (PII) is removed from notes before any text is sent to Amazon Bedrock.

The Al assistant must be able to detect PII in both English and Spanish and prevent the foundation model (FM) from returning medical identifiers in generated summaries. The Al assistant must detect PII and provide audit visibility for all redacted fields.

Which solution will meet these requirements?

A

Use Amazon Comprehend to detect PII in English and Spanish case notes and to apply automatic redaction before the Al assistant invokes Amazon Bedrock. Configure Amazon Bedrock guardrails to block disallowed medical identifiers in model outputs. Store redacted input and output logs in Amazon S3 for auditing.

B

Implement a multi-stage PII detection workflow that uses Amazon Comprehend to detect standard PII types. Implement a fine-tuned Amazon Bedrock model to detect specialty medical identifiers. Combine outputs to create a redacted version of each note and then invoke Amazon Bedrock to summarize the notes. Store audit logs in Amazon S3 to record redaction decisions.

C

Use Amazon SageMaker Clarify to detect PII before invoking the FM, Configure AWS WAF rules to block outbound responses that contain sensitive patient data, Store all processed case notes in Amazon DynamoDB.

D

Use AWS Glue DataBrew to apply a one-time masking transformation to all case notes. Send the masked notes to Amazon Bedrock and implement AWS Lake Formation column-level security to restrict downstream access.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 27

A healthcare company uses Amazon Bedrock to deploy an application that generates summaries of clinical documents. The application experiences inconsistent response quality with occasional factual hallucinations. Monthly costs exceed the company's projections by 40%. A GenAI developer must implement a near real-time monitoring solution to detect hallucinations, identify abnormal token consumption, and provide early warnings of cost anomalies. The solution must require minimal custom development work and maintenance overhead.

Which solution will meet these requirements?

A

Configure Amazon CloudWatch alarms to monitor InputTokenCount and OutputTokenCount metrics to detect anomalies. Store model invocation logs in an Amazon S3 bucket. Use AWS Glue and Amazon Athena to identify potential hallucinations.

B

Run Amazon Bedrock evaluation jobs that use LLM-based judgments to detect hallucinations. Configure Amazon CloudWatch to track token usage. Create an AWS Lambda function to process CloudWatch metrics. Configure the Lambda function to send usage pattern notifications.

C

Configure Amazon Bedrock to store model invocation logs in an Amazon S3 bucket. Enable text output logging. Configure Amazon Bedrock guardrails to run contextual grounding checks to detect hallucinations. Create Amazon CloudWatch anomaly detection alarms for token usage metrics.

D

Use AWS CloudTrail to log all Amazon Bedrock API calls. Create a custom dashboard in Amazon QuickSight to visualize token usage patterns. Use Amazon SageMaker Model Monitor to detect quality drift in generated summaries.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 28

A company is using AWS Lambda and REST APIs to build a reasoning agent to automate support workflows. The system must preserve memory across interactions, share the relevant agent state, and support event-driven invocation and synchronous invocation. The system must also enforce access control and session-based permissions.

Which combination of steps provides the MOST scalable solution? (Choose two.)

A

Use Amazon Bedrock AgentCore to manage memory and session-aware reasoning. Deploy the agent with built-in identity support, event handling, and observability.

B

Register the Lambda functions and the REST APIs as actions by using Amazon API Gateway and Amazon EventBridge. Enable Amazon Bedrock AgentCore to invoke the Lambda functions and the REST APIs without custom orchestration code.

C

Use Amazon Bedrock Agents for reasoning and conversation management. Use AWS Step Functions and Amazon SQS queues for orchestration. Store the agent state in Amazon DynamoDB to maintain memory between steps.

D

Deploy the reasoning logic as a container on Amazon ECS behind Amazon API Gateway. Use Amazon Aurora to store memory data and identity data.

E

Build a custom RAG pipeline by using Amazon Kendra and Amazon Bedrock. Use AWS Lambda to orchestrate tool invocations. Store the agent state in Amazon S3.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 29

An ecommerce company is developing a generative AI (GenAI) solution that uses Amazon Bedrock with Anthropic Claude to recommend products to customers. Customers report that some of the recommended products are not available for sale on the website or are not relevant to the customer. Customers also report that the solutions takes a long time to generate some recommendations.

The company investigates the issues and finds that most interactions between customers and the product recommendation solution are unique. The company confirms that the solutions recommends products that are not in the company's product catalog. The company must resolve these issues.

Which solution will meet this requirement?

A

Increase grounding within Amazon Bedrock Guardrails. Enable Automated Reasoning checks. Set up provisioned throughput.

B

Use prompt engineering to restrict the model responses to relevant products. Use streaming techniques such as the Invoke Model With Response Stream action to reduce perceived latency for the customers.

C

Create an Amazon Bedrock knowledge base. Implement Retrieval Augmented Generation (RAG). Set the Performance ConfigLatency parameter to optimized.

D

Store product catalog data in Amazon OpenSearch Service. Validate the model's product recommendations against the product catalog. Use Amazon DynamoDB to implement response caching.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 30

A company is building an AI advisory application by using Amazon Bedrock. The application will provide recommendations to customers. The company needs the application to explain its reasoning process and cite specific sources for data. The application must retrieve information from company data sources and show step-by-step reasoning for recommendations. The application must also link data claims to source documents and maintain response latency under 3 seconds.

Which solution will meet these requirements with the LEAST operational overhead?

A

Use Amazon Bedrock Knowledge Bases with source attribution enabled. Use the Anthropic Claude Messages API with RAG to set high-relevance thresholds for source documents. Store reasoning and citations in Amazon S3 for auditing purposes.

B

Use Amazon Bedrock with Anthropic Claude models and extended thinking. Configure a 4,000-token thinking budget. Store reasoning traces and citations in Amazon DynamoDB for auditing purposes.

C

Configure Amazon SageMaker AI with a custom Anthropic Claude model. Use the model's reasoning parameter and AWS Lambda to process responses. Add source citations from a separate Amazon RDS database.

D

Use Amazon Bedrock with Anthropic Claude models and chain-of-thought reasoning. Configure custom retrieval tracking with the Amazon Bedrock Knowledge Bases API. Use Amazon CloudWatch to monitor response latency metrics.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 31

A media company must use Amazon Bedrock to implement a robust governance process for AI-generated content. The company needs to manage hundreds of prompt templates. Multiple teams use the templates across multiple AWS Regions to generate content. The solution must provide version control with approval workflows that include notifications for pending reviews. The solution must also provide detailed audit trails that document prompt activities and consistent prompt parameterization to enforce quality standards.

Which solution will meet these requirements?

A

Configure Amazon Bedrock Studio prompt templates. Use Amazon CloudWatch to create dashboards that display prompt usage metrics. Store the approval status of content in Amazon DynamoDB. Use AWS Lambda functions to enforce approvals.

B

Use Amazon Bedrock Prompt Management to implement version control. Configure AWS CloudTrail for audit logging. Use IAM policies to control approval permissions. Create parameterized prompt templates by specifying variables.

C

Use AWS Step Functions to create an approval workflow. Store prompts as documents in Amazon S3. Use tags to implement version control. Use Amazon EventBridge to send notifications.

D

Deploy Amazon SageMaker Canvas with prompt templates that are stored in Amazon S3. Use AWS CloudFormation to implement version control. Use AWS Config to enforce approval policies.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 32

A company uses an organization in AWS Organizations with all features enabled to manage multiple AWS accounts. Employees use Amazon Bedrock across multiple accounts. The company must prevent specific topics and proprietary information from being included in prompts to Amazon Bedrock models. The company must ensure that employees can use only approved Amazon Bedrock models. The company centrally manages IAM roles for employees.

Which combination of solutions will meet these requirements? (Choose two.)

A

Create an IAM permissions boundary for each employee's IAM role. Configure the permissions boundary to require an approved Amazon Bedrock guardrail identifier to invoke Amazon Bedrock models. Create an SCP that allows employees to use only approved models.

B

Create an SCP that allows employees to use only approved models. Configure the SCP to require employees to specify a guardrail identifier in calls to invoke an approved model.

C

Create an SCP that prevents an employee from invoking a model if a centrally deployed guardrail identifier is not specified in a call to the model. Create a permissions boundary on each employee's IAM role that allows each employee to invoke only approved models.

D

Use AWS CloudFormation to create a custom Amazon Bedrock guardrail that has a block filtering policy. Use stack sets to deploy the guardrail to each account in the organization.

E

Use AWS CloudFormation to create a custom Amazon Bedrock guardrail that has a mask filtering policy. Use stack sets to deploy the guardrail to each account in the organization.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 33

An insurance company uses existing Amazon SageMaker AI infrastructure to support a web-based application that allows customers to predict what their insurance premiums will be. The company stores customer data that is used to train the SageMaker AI model in an Amazon S3 bucket. The dataset is growing rapidly. The company wants a solution to continuously re-train the model. The solution must automatically re-train and re-deploy the model to the application when an employee uploads a new customer data file to the S3 bucket.

Which solution will meet these requirements?

A

Use AWS Glue to run an ETL job on each uploaded file. Configure the ETL job to use the AWS SDK to invoke the Sage Maker AI model endpoint. Use real-time inference with the endpoint to re-deploy the model after it is re-trained on the updated customer dataset.

B

Create an AWS Lambda function and webhook handlers to generate an event when an employee uploads a new file. Configure SageMaker Pipelines to re-deploy the model after it is re-trained on the updated customer dataset. Use Amazon EventBridge to create an event bus. Set the Lambda function event as the source and SageMaker Pipelines as the target.

C

Create an AWS Step Functions Express workflow with AWS SDK integrations to retrieve the customer data from the S3 bucket when an employee uploads a new file to the S3 bucket. Use a SageMaker Data Wrangler flow to export the data from the S3 bucket to SageMaker Autopilot. Use SageMaker Autopilot to re-deploy the model after it has been re-trained on the updated customer dataset.

D

Create an AWS Step Functions Standard workflow. Configure the first state to call an AWS Lambda function to respond when an employee uploads a new file to the S3 bucket. Use a pipeline in SageMaker Pipelines to re-deploy the model after it has been re-trained on the updated customer dataset. Use the next state in the workflow to run the pipeline when the first state receives a response.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 34

A GenAI developer is building a Retrieval Augmented Generation (RAG)-based customer support application that uses Amazon Bedrock foundation models (FMs). The application needs to process 50 GB of historical customer conversations that are stored in an Amazon S3 bucket as JSON files. The application must use the processed data as its retrieval corpus. The application's data processing workflow must extract relevant data from customer support documents, remove customer personally identifiable information (PII), and generate embeddings for vector storage. The processing workflow must be cost-effective and must finish within 4 hours.

Which solution will meet these requirements with the LEAST operational overhead?

A

Use AWS Lambda and Amazon Comprehend to process files in parallel, remove PII, and call Amazon Bedrock APIs to generate vectors. Configure Lambda concurrency limits and memory settings to optimize throughput.

B

Create an AWS Glue ETL job to run PII detection scripts on the data. Use Amazon SageMaker Processing to run the HuggingFaceProcessor to generate embeddings by using a pre-trained model. Store the embeddings in Amazon OpenSearch Service.

C

Deploy an Amazon EMR cluster that runs Apache Spark with user-defined functions (UDFs) that call Amazon Comprehend to detect PII. Use Amazon Bedrock APIs to generate vectors. Store outputs in Amazon Aurora PostgreSQL with the pgvector extension.

D

Implement a data processing pipeline that uses AWS Step Functions to orchestrate a workload that uses Amazon Comprehend to detect PII and Amazon Bedrock to generate embeddings. Directly integrate the workflow with Amazon OpenSearch Serverless to store vectors and provide similarity search capabilities.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 35

A financial services company is creating a Retrieval Augmented Generation (RAG) application that uses Amazon Bedrock to generate summaries of market activities. The application relies on a vector database that stores a small proprietary dataset that has a low index count. The application must perform similarity searches. The Amazon Bedrock model's responses must maximize accuracy and maintain high performance.

The company needs to configure the vector database and integrate it with the application.

Which solution will meet these requirements?

A

Launch an Amazon MemoryDB cluster and configure the index by using the Flat algorithm. Configure a horizontal scaling policy based on performance metrics.

B

Launch an Amazon MemoryDB cluster and configure the index by using the Hierarchical Navigable Small World (HNSW) algorithm. Configure a vertical policy based on performance metrics.

C

Launch an Amazon Aurora PostgresSQL cluster and configure the index by using the Inverted File with Flat Compression (IVFFlat) algorithm. Configure the instance class to scale to a larger size when the load increases.

D

Launch an Amazon DocumentDB cluster that has an Inverted File with Flat Compression (IVFFlat) index and a high probe value. Configure connections to the cluster as a replica set Distribute reads to replica instances.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 36

A company uses Amazon Bedrock to build a Retrieval Augmented Generation (RAG) system. The RAG system uses an Amazon Bedrock knowledge base that is based on an Amazon S3 bucket as the data source for emergency news video content. The system retrieves transcripts, archived reports, and related documents from the S3 bucket.

The RAG system uses state-of-the-art embedding models and a high-performing retrieval setup. However, users report slow responses and irrelevant results, which cause decreased user satisfaction. The company notices that vector searches are evaluating too many documents across too many content types and over long periods of time.

The company determines that the underlying models will not benefit from additional fine tuning. The company must improve retrieval accuracy by applying smarter constraints. The company wants a solution that requires minimal changes to the existing architecture.

Which solution will meet these requirements?

A

Enhance embeddings by using a domain-adapted model that is specifically trained on emergency news content for improved vector similarity.

B

Migrate to Amazon OpenSearch Service. Use vector fields and metadata filters to define the scope of results retrieval.

C

Enable metadata-aware filtering within the Amazon Bedrock knowledge base by indexing S3 object metadata.

D

Migrate to an Amazon Q Business index to perform structured metadata filtering and document categorization during retrieval.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 37

An enterprise application uses an Amazon Bedrock foundation model (FM) to process and analyze 50 to 200 pages of technical documents. Users are experiencing inconsistent responses and receiving truncated outputs when processing documents that exceed the FM's context window limits.

Which solution will resolve this problem?

A

Configure fixed-size chunking at 4,000 tokens for each chunk with 20% overlap. Use application-level logic to link multiple chunks sequentially until the FM's maximum context window of 200,000 tokens is reached before making inference calls.

B

Use hierarchical chunking with parent chunks of 8,000 tokens and child chunks of 2,000 tokens. Use Amazon Bedrock Knowledge Bases built-in retrieval to automatically select relevant parent chunks based on query context. Configure overlap tokens to maintain semantic continuity.

C

Use semantic chunking with a breakpoint percentile threshold of 95% and a buffer size of 3 sentences. Use the Amazon Bedrock Retrieve And Generate API call to dynamically select the most relevant chunks based on embedding similarity scores.

D

Create a pre-processing AWS Lambda function that analyzes document token count by using the FM's tokenizer. Configure the lambda function to split documents into equal segments that fit within 80% of the context window. Configure the Lambda function to process each segment independently before aggregating the results.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 38

A financial services company needs to build a document analysis system that uses Amazon Bedrock to process quarterly reports. The system must analyze financial data, perform sentiment analysis, and validate compliance across batches of reports. Each batch contains 5 reports. Each report requires multiple foundation model (FM) calls. The solution must finish the analysis within 10 seconds for each batch. Current sequential processing takes 45 seconds for each batch.

Which solution will meet these requirements?

A

Use AWS Lambda functions with provisioned concurrency to process each analysis type sequentially. Configure the Lambda function timeouts to 10 seconds. Configure automatic retries with exponential backoff.

B

Use AWS Step Functions with a Parallel state to invoke separate AWS Lambda functions for each analysis type simultaneously. Configure Amazon Bedrock client timeouts. Use Amazon CloudWatch metrics to track execution time and model inference latency.

C

Create an Amazon SQS queue to buffer analysis requests. Deploy multiple AWS Lambda functions with reserved concurrency. Configure each Lambda function to process different aspects of each report sequentially and then combine the results.

D

Deploy an Amazon ECS cluster that runs containers that process each report sequentially. Use a load balancer to distribute batch workloads. Configure an auto-scaling policy based on CPU utilization to handle demand fluctuations.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 39

A company is building a generative AI (GenAI) application that produces content based on a variety of internal and external data sources. The company wants to ensure that the generated output is fully traceable. The application must support data source registration and enable metadata tagging to attribute content to its original source. The application must also maintain audit logs of data access and usage throughout the pipeline.

Which solution will meet these requirements?

A

Use AWS Lake Formation to catalog data sources and control access. Apply metadata tags directly in Amazon S3. Use AWS CloudTrail to monitor API activity.

B

Use AWS Glue Data Catalog to register and tag data sources. Use Amazon CloudWatch Logs to monitor access patterns and application behavior.

C

Store data in Amazon S3 and use object tagging for attribution. Use AWS Glue Data Catalog to manage schema information. Use AWS CloudTrail to log access to S3 buckets.

D

Use AWS Glue Data Catalog to register all data sources. Apply metadata tags to attribute data sources. Use AWS CloudTrail to log access and activity across services.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 40

Company configures a landing zone in AWS Control Tower. The company handles sensitive data that must remain within the European Union. The company must use only the eu-central-1 Region. The company uses SCPs to enforce data residency policies. GenAI developers at the company are assigned IAM roles that have full permissions for Amazon Bedrock.

The company must ensure that GenAI developers can use the Amazon Nova Pro model through Amazon Bedrock only by using cross-Region inference (CRI) and only in eu-central-1. The company enables model access for the GenAI developer IAM roles in Amazon Bedrock. However, when a GenAI developer attempts to invoke the model through the Amazon Bedrock Chat/Text playground, the GenAI developer receives the following error.

User: arn:aws:sts::123456789012:assumed-role/AssumedDevRole/DevUserName

Action: bedrock:InvokeModelWithResponseStream

On resource(s): arn:aws:bedrock:eu-west-3::foundation-model/amazon.nova-pro-v1:0

Context: a service control policy explicitly denies the action

The company needs a solution to resolve the error. The solution must retain the company's existing governance controls and must provide precise access control. The solution must comply with the company's existing data residency policies.

Which combination of solutions will meet these requirements? (Choose two.)

A

Add an AdministratorAccess policy to the GenAI developer IAM role.

B

Extend the existing SCPs to enable CRI for the eu.amazon.nova-pro-v1:0 inference profile.

C

Enable Amazon Bedrock model access for Amazon Nova Pro in the eu-west-3 Region.

D

Validate that the GenAI developer IAM roles have permissions to invoke Amazon Nova Pro through the eu.amazon.nova-pro.v1:0 inference profile on all European Union AWS Regions that can serve the model.

E

Extend the existing SCP to enable CRI for the eu.* inference profile.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 41

A company is designing an API for a generative AI (GenAI) application that uses a foundation model (FM) that is hosted on a managed model service. The API must stream responses to reduce latency, enforce token limits to manage compute resource usage, and implement retry logic to handle model timeouts and partial responses.

Which solution will meet these requirements with the LEAST operational overhead?

A

Integrate an Amazon API Gateway HTTP API with an AWS Lambda function to invoke Amazon Bedrock. Use Lambda response streaming to stream responses. Enforce token limits within the Lambda function. Implement retry logic for model timeouts by using Lambda and API Gateway timeout configurations.

B

Connect an Amazon API Gateway HTTP API directly to Amazon Bedrock. Simulate streaming by using client-side polling. Enforce token limits on the frontend. Configure retry behavior by using API Gateway integration settings.

C

Connect an Amazon API Gateway WebSocket API to an Amazon ECS service that hosts a containerized inference server. Stream responses by using the WebSocket protocol. Enforce token limits within Amazon ECS. Handle model timeouts by using ECS task lifecycle hooks and restart policies.

D

Integrate an Amazon API Gateway REST API with an AWS Lambda function that invokes Amazon Bedrock. Use Lambda response streaming to stream responses. Enforce token limits within the Lambda function. Implement retry logic by using Lambda and API Gateway timeout configurations.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 42

A social media company deploys an Amazon Bedrock application that generates article summaries for journalists. The application processes thousands of requests daily across multiple foundation models (FMs).

Costs for the application increase 40% over two weeks. The company needs a solution to identify which FMs, users, or usage patterns are driving the cost increase. The solution must track token usage for each FM, detect unusual consumption patterns, and alert the company when costs exceed set thresholds.

Which solution will meet these requirements?

A

Configure Amazon Bedrock FM invocation logging to Amazon S3. Develop custom scripts to parse logs and calculate token usage daily. Manually review token consumption reports to identify cost increases.

B

Collect Amazon CloudWatch metrics for input tokens and output tokens for each FM invocation. Configure CloudWatch anomaly detection on token usage metrics. Create CloudWatch alarms that activate when token usage exceeds dynamic thresholds.

C

Use AWS Cost Explorer to review Amazon Bedrock costs on a weekly schedule. Analyze spending trends by service and usage type. Manually identify which FMs are driving cost increases.

D

Use Amazon CloudWatch Application Signals to monitor application performance metrics. Configure composite alarms for invocation latency and error rates. Use AWS Cost Anomaly Detection to alert the company when Amazon Bedrock costs exceed set thresholds.

Premium Solution Locked

Unlock all 213 answers & explanations

QUESTION 43

A company is developing a new AI-powered application that needs to integrate with various specialized tools. These tools currently run as Model Context Protocol (MCP) servers on the local machines of developers and do not maintain states between invocations. The company plans to deploy each MCP server as an AWS Lambda function to support the company's production application.

The solution must be accessible to both internal applications and authorized third-party partners. The solution must use strict authentication and authorization controls.

Which additional steps will meet these requirements with the LEAST operational overhead?

A

Create a custom Lambda invocation transport by using the Lambda Invoke API. Implement IAM authentication and grant InvokeFunction permissions to authorized users and roles.

B

Expose the Lambda functions through Amazon API Gateway REST API endpoints. Implement API keys for authentication. Configure the applications that need to access the MCP servers to use standard HTTP requests instead of the MCP protocol.

C

Create Lambda function URLs and enable a custom Streamable HTTP transport and SigV4. Implement AWS IAM authentication. Grant InvokeFunctionUrl permissions to authorized users and roles.

D

Expose the Lambda function through Amazon API Gateway HTTP API endpoints with the Streamable HTTP transport. Use Amazon Cognito to implement OAuth authentication. Configure API Gateway to validate OAuth tokens.

Premium Solution Locked

Unlock all 213 answers & explanations

Amazon AWS Certified Generative AI Developer - Professional (AIP-C01)

Test the Practice Engine

Premium Bundle

Standard Simulation

Basic Tier

Verified Community

Success Network

Exam Overview

Skills Measured

Career Path

Common Questions

Is the material up to date?

What format do I get?

Is there a guarantee?

When do I get the download?

Free Study Guide Samples

Why other options are incorrect:

Full Question Bank Locked

Certification Path

Related Certifications

Customer Reviews

David M.

Sarah J.

Michael C.

Need Assistance?

Get Exam Updates