AIP-C01 Questions and Answers

Option C is the correct solution because it directly addresses both correctness and performance issues by grounding the model’s responses in authoritative product data using Retrieval Augmented Generation. Amazon Bedrock Knowledge Bases are designed to connect foundation models to trusted enterprise data sources, ensuring that generated responses are constrained to known, validated content.

By ingesting the product catalog into a knowledge base, the GenAI application retrieves only products that actually exist in the catalog. This prevents hallucinated or unavailable recommendations, which is a common issue when models rely solely on prompt instructions without retrieval grounding. RAG ensures that the model’s output is based on retrieved facts rather than learned generalizations.

Setting the PerformanceConfigLatency parameter to optimized enables Bedrock to prioritize lower-latency retrieval and inference paths, improving responsiveness for real-time recommendation scenarios. This directly addresses the reported performance issues without requiring provisioned throughput or caching strategies that are ineffective for mostly unique interactions.

Option A improves safety and latency predictability but does not ensure recommendations are limited to valid products. Option B relies on prompt constraints, which are not sufficient to prevent hallucinations. Option D introduces additional validation and caching layers but increases complexity and does not improve generation relevance.

Therefore, Option C best resolves both relevance and latency challenges using AWS-native, low-maintenance GenAI integration patterns.

Option B is correct because Amazon Bedrock Knowledge Bases supports structured document metadata that can be stored separately from the text and used as filters during retrieval. Case type, jurisdiction, and case outcome are ideal filterable metadata attributes, while author and version fields preserve provenance for returned documents. AWS documentation confirms that S3 knowledge-base data sources can use sidecar metadata files and that metadata remains available for filtering even when it is excluded from the embedding. This gives the application precise semantic retrieval plus structured constraints without custom extraction at query time. Option A lacks structured filtering. Option C adds unnecessary Comprehend processing. Option D embeds metadata inside document text, which does not provide reliable field-level filtering and adds unnecessary agent-side extraction complexity. AWS Documentation

================

Option B is the correct solution because Amazon Q Business natively supports access control lists (ACLs) for S3 data sources using a single, centralized JSON file that maps S3 prefixes to IAM Identity Center groups. This approach directly aligns with the company’s data organization model, where each department’s data is stored under a distinct S3 prefix and each employee belongs to exactly one department group.

Using a single acl.json file at the bucket root minimizes operational overhead by centralizing access control logic in one location. Administrators can update department mappings without touching individual folders or changing IAM permissions, which simplifies governance and reduces the risk of configuration drift. Amazon Q Business automatically evaluates the user’s IAM Identity Center group membership at query time and filters accessible documents accordingly.

Option A increases operational complexity by requiring a separate ACL file in every department folder, which becomes difficult to maintain as departments or prefixes change. Option C attempts to enforce access using IAM permissions sets, but Amazon Q Business access control for S3 data sources is not designed to be managed through IAM condition logic and would significantly increase complexity. Option D introduces a custom metadata structure that is not the supported mechanism for Amazon Q Business access enforcement.

Therefore, Option B provides the cleanest, most scalable, and AWS-recommended solution for enforcing department-based access control with the least operational effort.

Option C best satisfies the requirement for flexible FM testing across environments while minimizing operational complexity and aligning with AWS-recommended deployment practices. Amazon Bedrock supports invoking on-demand foundation models through the FoundationModel abstraction, which allows applications to dynamically reference different models without requiring dedicated provisioned capacity. This is ideal for experimentation and A/B testing in both development and production environments.

Using a single AWS CDK application ensures infrastructure consistency and reduces duplication. Environment-specific configuration, such as selecting different foundation model IDs, can be externalized through parameters, context variables, or environment-specific configuration files. This allows product owners to easily switch between FMs in each environment without modifying application logic.

A single AWS CodePipeline with distinct deployment stages for development and production is an AWS best practice for multi-environment deployments. It enforces consistent build and deployment steps while still allowing environment-level customization. AWS CodeBuild deploy actions enable automated, repeatable deployments, reducing manual errors and improving governance.

Option A increases complexity by introducing multiple pipelines and relies on provisioned models, which are not necessary for FM evaluation and experimentation. Provisioned throughput is better suited for predictable, high-volume production workloads rather than frequent model switching.

Option B creates unnecessary operational overhead by duplicating CDK applications and pipelines, making long-term maintenance more difficult.

Option D directly conflicts with infrastructure-as-code best practices by manually recreating development resources, which increases configuration drift and reduces reliability.

Therefore, Option C provides the most flexible, scalable, and AWS-aligned solution for testing and switching foundation models across development and production environments.

Option D is the best solution because it delivers a fully managed, scalable pipeline with minimal infrastructure management while meeting the 50 GB and 4-hour constraint. AWS Step Functions provides a serverless orchestration layer that can coordinate parallel processing steps, retries, and error handling without managing clusters or tuning long-running compute.

Using Amazon Comprehend for PII detection fulfills the requirement to remove customer PII in a managed and consistent way. Step Functions can coordinate Comprehend calls at scale and route sanitized outputs into the embedding step. Generating embeddings with Amazon Bedrock keeps the entire workflow within AWS managed services, eliminates the need to maintain custom embedding models, and supports consistent vector representations for downstream retrieval.

Direct integration with Amazon OpenSearch Serverless provides a low-operations vector store that can handle large-scale indexing and similarity search without cluster sizing, node maintenance, or shard management. This aligns strongly with the requirement for least operational overhead and supports growth beyond the initial 50 GB corpus. Step Functions can batch and parallelize ingestion into OpenSearch Serverless to meet the 4-hour completion goal in a cost-effective manner by controlling concurrency, chunk sizes, and failure handling.

Option A can be difficult and costly at this scale because Lambda concurrency and per-invocation overhead can become complex to tune for 50 GB within 4 hours. Option B introduces SageMaker Processing and embedding model management, increasing operational complexity. Option C requires EMR cluster management and tuning, which is the opposite of minimal overhead.

Therefore, Option D is the most operationally efficient, scalable, and managed approach to build the required PII-sanitized embedding pipeline for a RAG corpus.

The correct combination is A, D, and F because these guardrail features directly map to the stated financial compliance and governance requirements.

Option A is required because denied topics guardrails are explicitly designed to block entire categories of requests, such as requests for guaranteed returns or specific stock recommendations. These are regulatory-sensitive scenarios where partial filtering is insufficient and full blocking is required to prevent violations.

Option D is correct because custom word filters are the appropriate guardrail mechanism to block references to specific competitor names. Content filters are category-based (such as hate, sexual, or violence-related content) and are not suitable for blocking organization-specific competitor references. Custom word filters allow precise blocking at both input and output stages.

Option F is required because a high grounding score threshold enforces that model outputs must be strongly supported by approved source material. This prevents the AI assistant from making speculative or unfounded claims that are not aligned with the company’s approved financial guidance, which is critical in regulated financial environments.

Option B is incorrect because content filters do not target domain-specific financial advice patterns. Option C is incorrect for the same reason—competitor names are not a content filter category. Option E would weaken factual grounding and increase hallucination risk.

Therefore, A, D, and F together provide topic blocking, competitor exclusion, and factual grounding enforcement.

Option A best meets the requirements because it directly implements a Retrieval Augmented Generation pattern for conversational recommendations using managed Amazon Bedrock capabilities and a scalable vector store. The company’s source data already resides in Amazon S3, which aligns naturally with Amazon Bedrock Knowledge Bases ingestion workflows. A knowledge base can ingest book text, reviews, and metadata, generate embeddings using a supported embedding model, and persist those vectors in a purpose-built vector backend such as Amazon OpenSearch Service. This enables semantic retrieval that is well suited to conversation-driven intent, where user prompts are often descriptive and do not map cleanly to keyword filters.

The requirement to suggest books based on conversations implies the system must interpret natural language context and retrieve relevant passages, reviews, and metadata to ground the recommendation. Knowledge Bases provide managed orchestration for embedding creation and retrieval, which reduces development effort compared to building custom embedding pipelines. OpenSearch Service provides scalable vector search and k-nearest neighbors style similarity retrieval, which supports low-latency responses when properly indexed and sized.

For scaling to more than 10,000 concurrent users, the API layer design in option A is a common AWS pattern: Amazon API Gateway provides a managed front door with throttling and request handling, while AWS Lambda scales horizontally with demand and can invoke the knowledge base retrieval operations. This separates compute scaling from the vector store scaling and helps keep latency predictable under load.

Option B is not the best choice because DynamoDB is not the standard native vector store target for Amazon Bedrock Knowledge Bases in this context and would introduce additional implementation complexity around vector indexing and similarity search behavior. Option C requires substantial ML lifecycle work, model hosting, tuning, and continuous iteration to achieve quality recommendations at scale. Option D provides strong enterprise search, but it focuses on retrieval and FAQs rather than a managed RAG recommendation workflow grounded in embeddings and conversational context for generative responses.

Option A is the most comprehensive and architecturally aligned solution for meeting end-to-end data lineage, real-time PII filtering, and automated compliance reporting requirements in a medi cal GenAI system built on Amazon Bedrock. Each requirement maps directly to a managed AWS service that is purpose-built for governance, security, and compliance.

AWS Glue Data Catalog is designed to register datasets across multiple sources and maintain metadata that supports lineage tracking. By cataloging all inputs that flow into the Bedrock-based system, the organization can trace how data moves from ingestion through processing and storage, which is essential for regulatory audits in healthcare environments.

For real-time PII filtering , Amazon Bedrock Guardrails provide native PII detection and filtering during model inference. Guardrails operate inline with model invocation, ensuring sensitive information is blocked or redacted before responses are returned to users. This satisfies the requirement for real-time protection rather than post-processing analysis.

AWS CloudTrail delivers a complete audit trail of all Amazon Bedrock API calls, including InvokeModel requests and configuration changes. Storing these logs in Amazon S3 enables long-term retention and supports compliance audits. CloudTrail ensures traceability of who accessed the system, when, and how it was used.

To strengthen compliance monitoring, Amazon Macie continuously scans stored data for sensitive information and automatically classifies findings. Publishing Macie findings to Amazon CloudWatch Logs and visualizing them through dashboards enables near-real-time visibility into compliance posture and supports automated reporting workflows.

The other options fall short. Option B performs PII filtering at the application edge rather than at inference time and relies on scheduled analysis instead of real-time enforcement. Option C focuses on replication and document processing rather than inline GenAI governance. Option D uses services that are not designed for PII detection in text-based GenAI workflows and lacks native lineage tracking.

Therefore, A best fulfills all stated requirements using AWS-recommended governance and security capabilities.

Option B is the correct solution because it parallelizes independent foundation model inference tasks while maintaining orchestration, observability, and time-bound execution. AWS Generative AI best practices emphasize reducing end-to-end latency by parallelizing independent inference calls rather than scaling individual calls vertically.

In this scenario, each report requires multiple independent analyses such as financial extraction, sentiment analysis, and compliance validation. These tasks do not depend on each other’s output, making them ideal candidates for parallel execution. AWS Step Functions provides a Parallel state that can invoke multiple AWS Lambda functions simultaneously, drastically reducing total processing time compared to sequential execution.

By invoking Amazon Bedrock from separate Lambda functions in parallel, the system can reduce batch execution time from 45 seconds to well under the 10-second requirement, assuming each inference call remains within acceptable latency bounds. Step Functions also provide built-in error handling, retries, and state tracking, which improves reliability without increasing complexity.

CloudWatch metrics allow teams to monitor both workflow execution time and individual model inference latency, enabling performance tuning and operational visibility. Configuring client-side timeouts ensures that slow or failed model invocations do not block the entire batch.

Option A still processes tasks sequentially and therefore cannot meet the strict latency requirement. Option C introduces queuing delays and sequential processing within each report, which increases total execution time. Option D relies on container-based sequential processing and adds unnecessary operational overhead for a workload that is event-driven and latency-sensitive.

Therefore, Option B best meets the performance, scalability, and operational efficiency requirements for high-speed batch document analysis using Amazon Bedrock.

The combination of Options C and D delivers comprehensive, real-time observability for Amazon Bedrock workloads with the least operational overhead by relying on native integrations and managed services.

Amazon Bedrock publishes built-in CloudWatch metrics for model invocations and token usage. Option C leverages these native metrics directly, allowing teams to build centralized CloudWatch dashboards without additional data pipelines or custom processing. CloudWatch alarms provide threshold-based alerting for token consumption, enabling proactive cost and usage control across all foundation models. This approach aligns with AWS guidance to use native service metrics whenever possible to reduce operational complexity.

Option D complements CloudWatch by enabling advanced, stakeholder-specific visualizations through Amazon Managed Grafana. The zero-ETL integration allows Bedrock and CloudWatch metrics to be visualized directly in Grafana without building ingestion pipelines or managing storage layers. Grafana dashboards are particularly well suited for serving different audiences, such as engineering, finance, and product teams, each with customized views of token usage and trends.

Option A introduces unnecessary complexity by adding a business intelligence layer that is better suited for historical analytics than real-time operational monitoring. Option B is useful for deep log analysis but requires query maintenance and does not provide efficient real-time dashboards at scale. Option E involves multiple services and custom data flows, significantly increasing operational overhead compared to native metric-based observability.

By combining CloudWatch dashboards and alarms with Managed Grafana’s zero-ETL visualization capabilities, the company achieves real-time visibility, flexible dashboards, and automated alerting across all Amazon Bedrock foundation models with minimal operational effort.

To achieve the ultra-low latency requirement of less than 500 ms, the system must utilize streaming capabilities at every stage. Using Amazon Transcribe streaming with small (100-ms) audio chunks ensures that transcription begins immediately as the customer speaks. On the model side, Amazon Bedrock’s InvokeModelWithResponseStream allows the application to receive tokens as they are generated, rather than waiting for the entire completion, which is critical for real-time interactions. Amazon DynamoDB is the ideal choice for storing supervisor ratings during a live call because it provides the single-digit millisecond latency required for high-frequency writes without impacting the application ' s performance. Options involving batch processing or SQS queuing are unsuitable for sub-500 ms interactive requirements.

Amazon Bedrock AgentCore Gateway is specifically designed to expose existing APIs, Lambda functions, services, and MCP servers to agents through a standardized Model Context Protocol interface. AWS documents that Gateway can transform REST APIs represented by OpenAPI specifications into MCP-compatible tools and provide agents with a unified endpoint for tool discovery and invocation. This directly satisfies the requirement to add or expose warehouse, shipping, and customs capabilities through MCP without repeatedly modifying agent integration code.

AWS Step Functions is appropriate for deterministic workflow orchestration. Task states can represent individual reasoning or tool-execution stages, while built-in Retry configuration supports IntervalSeconds , MaxAttempts , and BackoffRate . These settings provide controlled retries with exponential backoff and a defined maximum retry budget rather than requiring bespoke retry code. AWS explicitly documents MaxAttempts as the mechanism that stops retry behavior after the configured limit is reached.

A Choice state supplies deterministic conditional routing. The shipment value can be evaluated and requests above $100,000 can be sent to an approval branch. The approval task can use the Step Functions callback/task-token integration, which pauses workflow execution until an external process—such as a human approval system—returns success or failure. AWS specifically identifies human approval as a callback-task use case.

D is weaker because critical retry and approval policy should not depend solely on probabilistic orchestration prompts. E and G introduce unnecessary custom operational logic. F misuses Guardrails: Bedrock Guardrails governs model input/output safety and policy filtering; it is not a general-purpose human workflow orchestrator for shipment authorization.

==========

Option C is the correct solution because it uses Amazon Bedrock cross-Region inference profiles, which are explicitly designed to support regional data residency, private connectivity, and resilience with minimal operational overhead.

By using a Europe-scoped inference profile, the application ensures that all inference requests are routed only within European Regions where the FM is deployed, such as eu-central-1 and eu-west-3. This satisfies data residency requirements while still providing resilience and load distribution across Regions.

Configuring an Amazon Bedrock VPC endpoint ensures that all traffic remains on the AWS private network. No public endpoints are used, which aligns with the company’s private networking requirements.

Extending existing SCPs to allow inference profile usage ensures that employees can access the FM only in approved Regions, maintaining governance across the Control Tower environment.

Options A and B introduce unnecessary custom routing layers and EC2 management. Option D moves away from Amazon Bedrock entirely and increases operational complexity.

Therefore, Option C is the only solution that satisfies private access, regional confinement, governance controls, and low operational overhead.

Option B is correct because the application is currently invoking the base foundation model identifier, which routes traffic to the on-demand capacity pool rather than the company’s purchased provisioned throughput. In Amazon Bedrock, provisioned throughput is attached to a specific provisioned resource created through the provisioned throughput APIs. To consume that reserved capacity, inference requests must target the provisioned resource identifier that represents the purchased throughput, not the generic model identifier used for on-demand inference.

The code snippet uses modelId= " anthropic.claude-v2 " . This value selects the on-demand endpoint for that model. As a result, requests are subject to on-demand quotas and throttling behavior, while the provisioned throughput remains idle. This directly explains the CloudWatch observation: provisioned capacity metrics show unused capacity because no traffic is being directed to the provisioned resource, and the on-demand path is throttling because it is exceeding the applicable on-demand limits during peak volume.

Replacing the modelId value with the provisioned throughput ARN returned by the CreateProvisionedModelThroughput workflow ensures the runtime invocation is routed to the reserved capacity. Once traffic is directed correctly, the purchased model units provide the consistent throughput required for predictable performance during business hours, which is exactly why provisioned throughput is used.

Option A could increase capacity, but it does not fix the core issue that the application is not using the provisioned resource at all. Option C can reduce the impact of throttling temporarily, but it adds latency and does not guarantee consistent throughput; it also still wastes the provisioned capacity. Option D changes the response delivery mechanism, but throttling is a capacity routing and quota issue, not a streaming API issue.

Option D requires the least development effort because it uses a managed retrieval workflow that bundles the most time-consuming parts of semantic search: embedding generation, vector indexing, and natural language retrieval. With an Amazon Bedrock knowledge base, the application does not need to implement and operate separate services to (1) generate embeddings for hundreds of millions of records, (2) store and manage vectors, (3) build query-time embedding conversion logic, and (4) implement k-NN search orchestration. Instead, the knowledge base is configured to automatically create embeddings during ingestion, and the application queries it using the Amazon Bedrock Retrieve API , which accepts natural language input and performs the vector search as a managed capability.

The performance requirement (95% of queries within 500 ms) is best served by a purpose-built vector search backend rather than running similarity search directly inside a transactional PostgreSQL system at this scale. A knowledge base is designed for retrieval patterns and can be backed by scalable vector stores, which helps meet latency goals under heavy concurrency. The hourly freshness requirement maps naturally to ingestion updates: the pipeline can re-ingest updated restaurant details on a schedule so the knowledge base remains current without building custom re-embedding workflows in application code.

Cost-effective scaling during peak periods is also easier with a managed retrieval layer because scaling the retrieval workload is separated from the operational database. This avoids overprovi sioning PostgreSQL for peak semantic-search traffic and reduces the engineering effort to tune performance, sharding, indexing, and retry logic.

Options B and C can work, but they require the team to build and maintain embedding pipelines, query embedding generation, vector index management, and operational scaling strategies. Option A does not provide semantic search because it relies on keyword-based matching rather than embeddings.

Option C best meets the requirements by aligning with AWS best practices for data isolation, access control, and scalable GenAI retrieval. Implementing a separate Amazon Bedrock knowledge base for each hotel ensures strict separation of data and permissions. This approach naturally enforces hotel-level access control without requiring complex policy logic or post-query filtering.

A multi-account structure further strengthens security and governance by isolating each hotel’s data plane. AWS recommends account-level isolation for workloads with strong tenancy or compliance boundaries. Hotel staff can be granted access only to their hotel’s account and corresponding knowledge base, eliminating the risk of cross-hotel data exposure.

Direct data ingestion into each knowledge base enables near real-time updates for critical data such as room availability. For information that does not change frequently, scheduled synchronization reduces ingestion cost while maintaining accuracy. This hybrid ingestion model balances freshness and operational efficiency.

Because Amazon Bedrock Knowledge Bases are fully managed, performance remains consistent during peak usage periods without the company managing indexing, scaling, or retrieval infrastructure. Each knowledge base scales independently, preventing noisy-neighbor issues that could arise in a centralized design.

Option A and B rely on a centralized knowledge base, which increases policy complexity and introduces risk of misconfigured access controls. Option D adds unnecessary orchestration complexity and does not inherently solve real-time data freshness requirements.

Therefore, Option C provides the most secure, scalable, and operationally efficient solution for enhancing the PMS with Amazon Bedrock Knowledge Bases.

Option B provides the least operational overhead because it keeps the solution primarily inside managed Amazon Bedrock capabilities, minimizing custom orchestration code and infrastructure to operate. The core requirements are domain grounding, reduced semantic dilution for complex questions, and consistent low-latency responses at high request volume. A Bedrock knowledge base is purpose-built for Retrieval Augmented Generation by ingesting domain documents, chunking content, generating embeddings, and retrieving the most relevant passages at runtime. This directly addresses the need to reference domain-specific medical terminology from authoritative documents to reduce ambiguity and improve factual accuracy.

Reducing semantic dilution typically requires improving the retrieval query so that the retriever focuses on the most relevant concepts, especially for long or multi-intent questions. Enabling query decomposition allows the system to break a complex medical query into smaller, more targeted sub-queries. This increases retrieval precision and recall for each sub-question, which helps the model generate a more accurate synthesized response grounded in the retrieved medical context.

Amazon Bedrock Flows provide a managed way to orchestrate multi-step generative AI workflows, such as preprocessing the input, performing retrieval against the knowledge base, invoking a foundation model, and formatting the final response. Because flows are managed, the com pany avoids maintaining custom state machines, multiple Lambda functions, or bespoke routing logic. This reduces operational overhead while still supporting repeatable, observable execution.

Compared with the alternatives, option A introduces an agent plus API Gateway routing and multiple model choices, increasing configuration and runtime complexity. Option C requires hosting and scaling custom models on SageMaker AI, which adds significant operational burden and latency risk. Option D relies on multiple Lambda functions orchestrated by an agent, which adds more moving parts and increases cold-start and integration overhead. Option B most directly meets the requirements with the smallest operational footprint.

Option B is the best solution because hierarchical chunking is specifically designed to preserve broader semantic context while still enabling precise retrieval at paragraph or sub-paragraph granularity. The problem described—answers citing irrelevant sections when a query spans multiple paper sections—often occurs when chunks are either too small (losing cross-paragraph context) or too “flat” (retrieving isolated snippets without their surrounding rationale).

In a scientific paper, related information is frequently distributed across methodology, results, and discussion. Flat, fixed-size chunking (Option A) can split these logically connected ideas into separate chunks, causing retrieval to surface fragments that match a term but not the full intent. Semantic chunking (Option C) improves boundary placement, but it does not inherently provide a multi-resolution structure that helps preserve section-level continuity at massive scale.

Hierarchical chunking solves this by creating parent chunks (larger context windows) that capture broader section context and child chunks (smaller units) that retain retrieval precision. When the retriever identifies relevant child chunks, it can also bring in the associated parent context so the foundation model sees the surrounding methodological or discussion framing. The defined overlaps further reduce the risk that key transitions or references are split across chunks.

This approach is well suited for a corpus of 25 million papers because it improves relevance without requiring a custom reranking model or a manual preprocessing pipeline. It remains operationally efficient because it is configured at the knowledge base level rather than implemented through custom code per document.

Option D introduces high operational complexity and inconsistent document handling at scale. Therefore, Option B best meets the requirement to preserve semantic context across related paragraphs and improve citation relevance across scientific paper sections.

Option C is the correct solution because it uses a single, well-tuned Amazon Bedrock guardrail that applies different actions to different content types, which is the recommended approach for minimizing false positives while enforcing strong policy controls.

Setting content filters to medium rather than high reduces overblocking of benign customer conversations while still preventing harmful content. Amazon Bedrock guardrails are designed to balance precision and recall, and medium sensitivity is commonly recommended for customer-facing financial services use cases.

Denied topics explicitly prevent the assistant from discussing investment advice, which is a regulatory requirement. Including definitions and sample phrases improves detection accuracy and reduces ambiguity.

Sensitive information filters support different actions per context. Masking PII in responses preserves conversational usefulness for legitimate customer support while preventing exposure of sensitive data. Blocking sensitive financial information in inputs prevents downstream processing of disallowed content before it reaches the foundation model.

Critically, enabling both input and output evaluation ensures that guardrails are applied consistently at every stage of interaction. Custom blocked messages and audit logging provide clear compliance evidence for regulators and internal audits.

Option A causes excessive false positives by blocking all PII outright. Option B introduces unnecessary complexity and is not how Bedrock guardrails are intended to be applied. Option D uses orchestration logic that Bedrock guardrails already handle natively.

Therefore, Option C best satisfies enforcement, flexibility, auditability, and accuracy requirements.

Option A is the correct solution because it relies on native Amazon Bedrock capabilities to deliver transparency, auditability, scalability, and low latency with minimal operational overhead. Amazon Bedrock Knowledge Bases provide a fully managed Retrieval Augmented Generation (RAG) implementation that automatically handles document ingestion, embedding, retrieval, and source attribution, enabling the application to cite authoritative content without building custom pipelines.

Enabling tracing for Amazon Bedrock Agents provides end-to-end visibility into agent reasoning steps, tool usage, and model interactions. This satisfies the requirement for comprehensive audit trails and supports regulatory review in financial services environments. Structured prompts further ensure that responses explicitly present reasoning and supporting evidence in a controlled, auditable format.

Using Amazon API Gateway and AWS Lambda allows the application to scale automatically to thousands of concurrent users without capacity planning. These services are designed for bursty workloads and can easily support the stated requirement of up to 10,000 concurrent users. Amazon CloudFront reduces latency by caching and accelerating content delivery, helping the application meet the strict 2-second response-time requirement.

Option B introduces a custom RAG pipeline with OpenSearch, increasing operational complexity and maintenance effort. Option C lacks native RAG integration and does not provide transparent reasoning or citation management. Option D focuses on offline compliance reporting rather than real-time transparency and low-latency responses.

Therefore, Option A best meets all requirements while minimizing infrastructure and operational overhead.

Option C is the best answer because it implements a decoupled, queue-based moderation pipeline that invokes the expensive model only when the low-latency model is not confident enough. Amazon SQS is designed to decouple distributed application components and support asynchronous processing. AWS documentation describes SQS as a fully managed message queuing service that enables decoupling and scaling of microservices, distributed systems, and serverless applications. This matches the requirement for high resiliency and sudden demand spikes because incoming requests can be buffered in a durable queue rather than overwhelming the model-processing layer.

Using AWS Fargate to process queue messages is also appropriate because Fargate provides serverless container compute for Amazon ECS or Amazon EKS workloads. It allows the company to run scalable processing workers without managing EC2 capacity directly. AWS Prescriptive Guidance includes architectures that use API Gateway, Amazon SQS, and AWS Fargate to process events asynchronously, which supports the same decoupled processing model required in this question.

The two-stage queue design also optimizes cost. The small, low-latency classifier is used first for all requests. Only requests with confidence below 0.65 are placed into the second queue and processed by the larger model. This avoids running the larger model for every moderation request. High-confidence results can be completed quickly by the first-stage processor, while uncertain results are isolated into an asynchronous second-stage workflow.

Option A is incorrect because it synchronously calls the larger model for low-confidence results, which violates the requirement to process low-confidence requests asynchronously. Option B is incorrect because it invokes both models for every request, increasing cost and eliminating the benefit of confidence-based escalation. Option D requires managing EC2 instances and uses keyword heuristics instead of model confidence, so it is less resilient and less aligned with Bedrock-based moderation. Therefore, option C is correct.

Option B is the correct evaluation configuration because it enables end-to-end assessment of both retrieval and generation quality while supporting direct comparison of chunking strategies and foundation models. Amazon Bedrock evaluation jobs are designed to support RAG workflows by evaluating how well retrieved context supports accurate and high-quality model outputs.

A retrieve-and-generate evaluation job evaluates the complete RAG pipeline, not just retrieval. This is essential for medical information use cases, where both the relevance of retrieved content and the correctness of generated responses directly impact user safety and trust. Including multiple chunking strategies in the evaluation dataset allows side-by-side comparison under identical prompts and conditions.

Custom precision-at-k metrics measure how effectively the retrieval component surfaces relevant chunks, while an LLM-as-a-judge metric provides qualitative scoring of generated responses. Using a numeric scale enables consistent, repeatable evaluation and supports automated quality gates. Amazon Bedrock supports LLM-based evaluators to score dimensions such as accuracy, completeness, and relevance.

Using the same evaluator model to assess outputs from both FMs ensures consistent scoring and eliminates evaluator bias. This configuration allows the company to define quantitative thresholds that must be met before deployment, enabling automated promotion through CI/CD pipelines.

Option A evaluates retrieval only and cannot assess generation quality. Option C introduces manual review, which does not scale and delays deployment. Option D separates retrieval and generation evaluation, making it harder to correlate chunking strategies with final output quality.

Therefore, Option B best meets the requirements for systematic evaluation, comparison, and quality enforcement in an Amazon Bedrock–based RAG system.

Option B is the correct solution because it delivers end-to-end governance, security, and auditability across Amazon Bedrock, Amazon Rekognition, and the underlying data layer while meeting strict privacy and compliance requirements.

Using IAM attribute-based access control (ABAC) allows the company to control access to foundation models and data based on department, role, or workload attributes rather than static permissions. This is critical for controlling FM access patterns at scale. Enforcing specific ModelId and GuardrailIdentifier values with IAM condition keys ensures that only approved models and guardrails are used, which directly supports secure model I/O and governance requirements.

Configuring VPC endpoints for Amazon Bedrock ensures that all model invocations remain on private AWS network paths, reducing data exfiltration risk and supporting privacy standards. AWS CloudTrail captures both management and data events, providing a definitive audit trail of who accessed which resources and when. Sending logs to CloudTrail Lake enables centralized, long-term, queryable auditing across services.

Amazon S3 server access logging adds file-level visibility into video archive access, which is essential for compliance and forensic analysis. Amazon CloudWatch alarms provide near real-time detection of anomalous or unauthorized activity across Amazon Bedrock, Amazon Rekognition, and AWS KMS.

Option A focuses primarily on model-level tracing but lacks comprehensive IAM governance and S3 access auditing. Option C provides partial controls but lacks identity-aware auditing and model governance. Option D focuses on anomaly detection and classification but does not explicitly control FM access patterns.

Therefore, Option B best satisfies all stated requirements in a unified, auditable, and security-first architecture.

Option D is the correct solution because it directly satisfies all three core requirements: data source registration, metadata-based attribution, and end-to-end audit logging, while remaining service-agnostic and scalable across internal and external data sources.

The AWS Glue Data Catalog is the AWS-native service for registering datasets and managing metadata centrally. It supports structured registration of diverse data sources and enables consistent tagging that can be used to attribute generated content back to its original source. This is essential for GenAI applications that combine multiple datasets and must provide traceability for outputs.

Metadata tags applied within the Glue Data Catalog ensure a consistent attribution framework that downstream systems—such as Retrieval Augmented Generation (RAG) pipelines or evaluation systems—can reference without embedding attribution logic directly in application code. This improves maintainability and governance.

AWS CloudTrail provides immutable audit logs of API activity across AWS services, including data access, metadata changes, and pipeline interactions. CloudTrail logs are critical for compliance and regulatory review because they capture who accessed which data, when, and through which service. This satisfies the requirement to maintain audit logs “throughout the pipeline,” not just at storage or application layers.

Option A introduces Lake Formation, which is primarily intended for fine-grained data lake permissions and is not required solely for traceability. Option B relies on CloudWatch Logs, which does not provide authoritative audit logging across services. Option C limits audit scope to S3 access and does not register or govern all data sources comprehensively.

Therefore, Option D provides the most complete and least intrusive solution for traceable, auditable GenAI data pipelines.

Option C is the correct solution because AWS AppConfig is purpose-built to manage dynamic application configurations with low latency, strong validation, and minimal operational overhead, which directly matches the company’s requirements.

AWS AppConfig enables the company to centrally manage model selection logic, inference parameters, and customer-tier routing rules without redeploying Lambda functions. By using feature flags, the company can easily perform A/B testing of new models or prompt strategies by gradually rolling out changes to a subset of users or customer tiers. This allows experimentation and controlled releases without code changes.

AppConfig also supports JSON schema validation, which is critical for validating parameters such as temperature, maximum token limits, and other model-specific settings before they are applied. This prevents invalid or unsafe configurations from being deployed and reduces the risk of runtime errors or degraded model behavior in production.

Using the AWS AppConfig Agent allows Lambda functions to retrieve configurations efficiently with built-in caching and polling mechanisms, minimizing latency and avoiding excessive calls to configuration services. This approach scales well for high-throughput, low-latency applications such as GenAI APIs behind Amazon API Gateway.

Option A introduces unnecessary redeployment logic and polling complexity. Option B requires building and maintaining custom configuration access patterns in DynamoDB and does not natively support feature flags or schema validation. Option D adds operational overhead by requiring ElastiCache cluster management and custom validation logic.

Therefore, Option C provides the most scalable, flexible, and low-maintenance solution for dynamic model switching, A/B testing, and safe configuration management in a GenAI application.

Option B is the only solution that satisfies all strict real-time, streaming, and latency requirements. Amazon Transcribe streaming with partial results allows transcription fragments to be delivered before the speaker finishes a sentence. This significantly reduces perceived latency and enables downstream processing to begin immediately, which is essential for maintaining sub-1-second end-to-end response times.

Using Amazon Bedrock’s InvokeModelWithResponseStream API enables token-level or chunk-level streaming responses from the foundation model. This allows the GenAI assistant to begin delivering suggestions to call center agents incrementally instead of waiting for a full model response. This streaming inference capability is critical for interactive, real-time agent assistance use cases.

Amazon API Gateway WebSocket APIs provide fully managed, bidirectional communication between backend services and agent dashboards. This ensures that updates flow continuously to agents as new transcription fragments and model outputs become available, preserving real-time responsiveness without requiring custom socket infrastructure.

Option A introduces additional synchronous processing layers and storage writes that increase latency. Option C uses batch transcription and post-call processing, which cannot meet real-time requirements. Option D uses embeddings and asynchronous messaging, which are not suitable for live incremental suggestions and bidirectional streaming.

Therefore, Option B best aligns with AWS real-time GenAI architecture patterns by combining streaming transcription, streaming model inference, and managed bidirectional communication while maintaining low latency and operational simplicity.

To handle high concurrency (100+ requests) with sub-500 ms response times and diverse authentication methods without infrastructure changes, Amazon API Gateway with Lambda authorizers is the optimal choice. The Lambda authorizers can evaluate multiple authentication tokens or signatures centrally before the request reaches the processing logic, preventing unauthorized traffic and potential authentication failures at scale. AWS Lambda integrated with Amazon Bedrock provides the scalability to handle ticket surges (up to 250,000 daily) without over-provisioning resources. For high availability (99.9%) and multi-region resilience, storing the resulting sentiment and responses in Amazon DynamoDB global tables ensures that data is accessible across regions with minimal latency. Option B is less secure due to the " NONE " auth type, and Option C introduces queuing latency that may exceed the 500 ms target.

Options A, D, and E provide the required capabilities while minimizing custom infrastructure. Amazon Transcribe converts speech in uploaded media into text, Amazon Comprehend provides managed sentiment detection, and Bedrock Knowledge Bases can support embedding-based semantic retrieval over transcript content. DynamoDB is appropriate for lightweight, scalable counters such as views and positive-feedback totals and avoids operating a relational vector database solely for metrics. The Strands Agents SDK provides prebuilt HTTP-request tooling for invoking the existing chat API and supports tool-based and multi-agent orchestration patterns, reducing bespoke integration code. Option B adds Step Functions and Lambda when Strands can coordinate the required application interactions directly. Option C requires custom orchestration and error-management code. Option F introduces Aurora PostgreSQL and pgvector even though vector retrieval is already handled through the Bedrock knowledge-base layer. AWS Documentation

================

Option C is the correct solution because it resolves throttling while preserving performance and avoiding fixed costs during low-traffic periods. Amazon Bedrock supports on-demand inference with usage-based pricing, making it well suited for applications with time-zone–dependent traffic spikes.

Throttling during peak hours typically occurs when inference requests exceed available regional capacity. Cross-Region inference allows Amazon Bedrock to automatically distribute requests across multiple AWS Regions, reducing contention and preventing throttling without requiring reserved or provisioned capacity. This approach ensures continuous operation while maintaining low latency for users in different geographic locations.

Invocation logging and native metrics such as InvocationThrottles, InputTokenCount, and OutputTokenCount provide visibility into usage patterns and capacity constraints. Monitoring these metrics enables teams to validate that traffic distribution is working as intended and that performance remains consistent during peak periods.

Option A introduces fixed hourly costs by relying on provisioned throughput, which directly violates the requirement to avoid unnecessary spend during low-traffic periods. Option B introduces regional failover complexity and reactive behavior instead of proactive load distribution. Option D does not address the root cause of throttling, as distributing traffic across model versions within the same Region does not increase available capacity.

Therefore, Option C best aligns with AWS Generative AI best practices for scalable, cost-efficient, global serverless applications.

Option D is the correct solution because Amazon Bedrock Knowledge Bases natively support reranking by using Amazon-managed reranker models, which are specifically designed to improve contextual relevance after the initial vector retrieval step. This approach directly addresses the root cause of the issue: semantically similar but contextually irrelevant documents being passed to the foundation model.

By enabling the reranking configuration within Amazon Bedrock Knowledge Bases, the application can automatically reorder retrieved documents based on deeper contextual understanding, such as regulatory scope, legal applicability, and semantic intent. This significantly improves retrieval precision, which reduces hallucinations and improves the factual accuracy of generated regulatory guidance.

Option D requires no additional infrastructure, no custom orchestration logic, and no separate model hosting. The reranking is fully managed by Amazon Bedrock and integrates seamlessly with the existing RetrieveAndGenerateStream workflow. This makes it the lowest operational overhead solution.

Option A introduces operational complexity by requiring a custom SageMaker endpoint, API Gateway routing, and model lifecycle management. Option B combines multiple unrelated services and introduces significant complexity without being purpose-built for RAG relevance ranking. Option C improves relevance but requires explicitly calling the Rerank API and modifying the application pipeline, which increases operational and integration effort compared to built-in reranking.

Therefore, Option D provides the most efficient, scalable, and AWS-recommended method to improve RAG retrieval quality while minimizing operational burden.

Option C most completely satisfies the mandatory controls. A model-independent fairness classifier can evaluate every generated response at inference time, enforce a blocking threshold, preserve classifier scores and associated metadata for auditability, and route questionable cases for human compliance review without retraining the underlying FM. Option B provides automated blocking but omits the stronger audit metadata and regulatory-review workflow explicitly required by the scenario. Option A is insufficient because SageMaker Clarify can monitor bias and publish bias metrics, but scheduled offline or periodic analysis does not provide a non-bypassable control on every Bedrock response. Option D is inappropriate because AWS WAF matching operates primarily on HTTP request components and pattern-based conditions; detecting prohibited words is not equivalent to evaluating demographic fairness or semantic bias in generated investment guidance.

Option A is the correct solution because it directly addresses all identified failure modes while preserving the existing Step Functions–based orchestration architecture with minimal redesign.

Using Step Functions Map states enables parallel execution of secret resolution workflows, which improves refresh latency for short-lived and expiring secrets. This ensures that secrets are refreshed in time before downstream agents require them. Passing updated secret metadata through Lambda outputs guarantees that subsequent steps always consume the latest resolved values, preventing agents from using stale data even after drift alerts are generated.

Versioning prompt flows in AWS AppConfig is critical to resolving cascading failures caused by outdated templates. AppConfig natively supports version control, validation, staged rollout, and rollback of configuration artifacts. By gating prompt flows through AppConfig, the company can immediately roll back faulty templates and prevent agents from reusing outdated instructions.

This solution maintains clear separation of concerns: Step Functions handle orchestration and parallelism, Lambda handles secret retrieval and metadata propagation, and AppConfig governs prompt lifecycle management. No additional event pipelines or custom retry coordination layers are required.

Option B oversimplifies the architecture and does not address secret lifecycle or drift. Option C introduces event-driven ordering complexity without solving prompt versioning. Option D introduces unnecessary tooling and dynamic prompt generation risk.

Therefore, Option A best resolves performance, correctness, and stability issues while minimizing operational overhead.

The correct answers are A and D because they collectively satisfy the requirements for structured inputs, consistent output formatting, and non-blocking detection of bullying language.

Option A is required because Amazon Bedrock Prompt Management enables prompt templates with explicit input variables and defined output formats. By defining the four required inputs as variables, the company ensures that every invocation of the Amazon Nova Pro FM receives the correct structured inputs. Specifying the output format ensures consistent responses, which is essential for a grants application workflow. Adding the managed prompt to the prompts node of the flow allows Bedrock Flows to invoke the model using this standardized configuration.

Option D addresses the requirement to receive notifications when bullying language is detected without blocking responses. Amazon Bedrock guardrails support content filters with configurable actions. By applying the insults content filter and setting the response action to detect, the system flags responses containing bullying or insulting language while still allowing the response to be returned. This enables monitoring, alerting, and auditing without interrupting application functionality.

Option B is incorrect because setting the filter response to block contradicts the requirement not to block all flagged responses. Option C introduces a prompt router, which is unnecessary because the application uses a single Amazon Nova Pro FM. Option E incorrectly attempts to enforce input variables and output formatting through an inference profile, which does not provide prompt-level variable enforcement or formatting guarantees.

Therefore, A and D together provide structured prompt management and non-blocking safety detection with minimal operational complexity.

Option B best satisfies the requirements with the least custom development effort by using native Amazon Bedrock capabilities for prompt experimentation, traffic management, fairness monitoring, and alerting. Amazon Bedrock Prompt Management allows teams to define and manage multiple prompt variants without code changes, making it ideal for comparing recommendation strategies across demographic groups.

Amazon Bedrock Flows enables controlled traffic allocation between prompt variants, which supports real-time A/B testing. This allows the company to collect live fairness metrics under production conditions instead of relying on offline analysis. Because Flows are fully managed, they eliminate the need for custom routing or experimentation frameworks.

Amazon Bedrock guardrails provide built-in monitoring and intervention mechanisms. When configured for fairness-related checks, guardrails can detect policy violations and surface metrics such as InvocationsIntervened, which indicate when outputs are modified or blocked due to rule enforcement. These metrics integrate directly with Amazon CloudWatch, enabling real-time dashboards and threshold-based alarms. Setting an alarm at a 15% discrepancy threshold satisfies the alerting requirement with minimal configuration.

Weekly reporting can be generated from CloudWatch metrics using scheduled exports or dashboards without building custom analytics pipelines. Option A requires significant custom post-processing logic. Option C introduces an additional service with higher operational overhead and is not optimized for real-time monitoring. Option D focuses on offline evaluation jobs and does not provide continuous real-time fairness monitoring.

Therefore, Option B provides the most AWS-native, scalable, and low-effort solution for fairness evaluation and monitoring.

Option D is the best fit because it implements a reliable event-driven MLOps workflow that automates retraining and redeployment with clear orchestration, auditability, and production-grade error handling. The requirement is explicit: whenever a new file is uploaded to Amazon S3, the system must retrain and then redeploy the model used by a web application. A common AWS pattern is to use an S3 event notification to trigger an AWS Lambda function, which then starts a controlled workflow. In option D, Lambda serves as the event handler that reacts immediately to the S3 upload event and passes the necessary context (bucket, object key, dataset version) into an AWS Step Functions Standard state machine.

Step Functions Standard is appropriate for model retraining pipelines because training and deployment steps can be long-running and benefit from durable state, retries, and failure handling. It provides execution history, making it easier to troubleshoot why a particular retraining run failed and to prove which dataset version produced which model version. This operational visibility is critical when the dataset is “growing rapidly” and retraining is frequent.

Within the workflow, Amazon SageMaker Pipelines is the right service to run the ML lifecycle stages in a repeatable way: data processing (if needed), training, evaluation/quality checks, model registration, and deployment to an endpoint used by the application. SageMaker Pipelines is purpose-built for CI/CD-style ML, supporting automated redeployments when a new approved model artifact is produced. By calling a pipeline execution from Step Functions, the company can add governance gates (for example, only deploy if evaluation metrics meet thresholds), and can apply consistent rollback and notification steps when deployment fails.

The other options are weaker: A confuses inference with retraining and does not provide deployment orchestration. B adds unnecessary webhook complexity and describes an awkward event bus configuration. C introduces Autopilot/Data Wrangler, which may be useful but adds extra moving parts and is not required to meet the trigger-and-redeploy requirement.

Option B is the correct solution because it leverages native Amazon Bedrock intelligent prompt routing, which is specifically designed to reduce cost and complexity in multi-model GenAI architectures. Intelligent prompt routing automatically analyzes incoming prompts and selects the most appropriate foundation model based on prompt characteristics and complexity—without requiring custom classification logic or orchestration code.

This approach directly meets the requirement for least implementation effort. The company does not need to deploy additional Lambda functions, maintain routing rules, or manage separate classification stages. Routing decisions are handled by Bedrock, which simplifies architecture and reduces operational risk.

By routing the majority (70%) of simple product inquiries to smaller, lower-cost models, the company minimizes inference cost and latency. More complex return policy inquiries are automatically routed to larger models that provide better reasoning capabilities, preserving response quality and customer satisfaction.

Because routing is handled inline by Bedrock, response latency remains low compared to multi-stage architectures that require an additional classification model call before inference. This is critical for customer service scenarios where responsiveness directly impacts satisfaction.

Option A introduces additional inference steps and custom logic. Option C increases cost by overusing a mid-sized model for all queries. Option D relies on brittle keyword rules and increases operational overhead through endpoint management.

Therefore, Option B delivers the optimal balance of cost efficiency, performance, and simplicity for dynamic model selection in Amazon Bedrock.

Option B is correct because the retrieval problem requires filtering before semantic similarity search, not merely returning metadata after retrieval. Amazon Bedrock Knowledge Bases can apply filters to metadata associated with source documents, so timestamp, author, and research-domain attributes can restrict the candidate set before vector results are returned. This prevents old, wrong-domain, or unauthorized-author documents from entering the retrieved context. AWS also documents S3 knowledge-base metadata files and filterable metadata attributes. Option A only returns metadata with already-selected vector results, so irrelevant documents can still be retrieved. Option C introduces DynamoDB and a two-stage custom retrieval path. Option D adds Comprehend and Aurora, increasing operational complexity. The key design principle is to make the required attributes filterable in the vector retrieval layer. AWS Documentation

================

Option D is the correct solution because it directly addresses all three requirements: high retrieval accuracy, hallucination detection, and reduced human review costs. AWS recommends a layered evaluation strategy for high-stakes domains such as healthcare, where generative outputs must be both accurate and safe.

Using an automated LLM-as-a-judge evaluation enables scalable, consistent assessment of generated responses for factual grounding, relevance, and hallucination risk. This automated screening significantly reduces the number of responses that require manual inspection. Only responses that fall below defined quality thresholds or exhibit ambiguous behavior are escalated to targeted human reviews, which optimizes review effort and cost.

The use of Amazon Bedrock built-in evaluations provides standardized metrics specifically designed for RAG systems, including retrieval precision, faithfulness to source documents, and hallucination rates. These evaluations integrate directly with Amazon Bedrock knowledge bases and models, eliminating the need to build and maintain custom evaluation pipelines.

Option A focuses on entity extraction confidence, which does not reliably detect hallucinations in generative text. Option B requires maintaining and scaling a separate fine-tuned evaluation model, increasing complexity and cost. Option C is useful for regression testing but cannot detect hallucinations in real-world, open-ended clinical queries.

Therefore, Option D provides the most effective and operationally efficient approach to maintaining clinical-grade accuracy while minimizing human review effort.

Option D is the intended Bedrock prompt-engineering solution. AWS guidance recommends using output indicators to explicitly describe constraints and characteristics that the model should follow when constructing its response. The prompt should clearly state that the promotional message is mandatory, where it should appear, and how it should be integrated with the surrounding recommendation. This gives the model a stronger output constraint than an indirect instruction embedded elsewhere in a dynamically assembled prompt. Option A uses Guardrails, which are designed to evaluate and constrain unsafe or disallowed content rather than require insertion of specific marketing text. Option B adds unnecessary model calls and still does not enforce the desired response structure. Option C introduces additional AgentCore and reranking processing for a requirement that should be addressed directly through explicit output instructions in the prompt. AWS Documentation

================

Option C directly addresses the root cause of truncated and inconsistent responses by using AWS- recommended semantic chunking and dynamic retrieval rather than static or sequential chunk processing. Amazon Bedrock documentation emphasizes that foundation models have fixed context windows and that sending oversized or poorly structured input can lead to truncation, loss of context, and degraded output quality.

Semantic chunking breaks documents based on meaning instead of fixed token counts. By using a breakpoint percentile threshold and sentence buffers, the content remains coherent and semantically complete. This approach reduces the likelihood that important concepts are split across chunks, which is a common cause of inconsistent summarization results.

The RetrieveAndGenerate API is designed specifically to handle large documents that exceed a model’s context window. Instead of forcing all content into a single inference call, the API generates embeddings for chunks and dynamically selects only the most relevant chunks based on similarity to the user query. This ensures that the FM receives only high-value context while staying within its context window limits.

Option A is ineffective because chaining chunks sequentially does not align with how FMs process context and risks exceeding context limits or introducing irrelevant information. Option B improves structure but still relies on larger parent chunks, which can lead to inefficiencies when processing very large documents. Option D processes segments independently, which often causes loss of global context and inconsistent summaries.

Therefore, Option C is the most robust, AWS-aligned solution for resolving truncation and consistency issues when processing large technical documents with Amazon Bedrock.

Option A is the most appropriate design because it provides scalable multi-agent orchestration, clear domain separation, and strong governance with minimal operational complexity. A supervisor-agent pattern is a standard AWS-recommended approach for multi-agent systems: one agent performs intent classification and routing, while specialized agents handle domain-specific tasks.

Isolating data with separate knowledge bases ensures that each specialized collaborator agent retrieves only the information relevant to its department. This improves response accuracy, reduces hallucinations, and supports privacy controls because clinical content, claims content, and scheduling content can have different access policies. IAM-based filtering ensures that each agent has permission only to the knowledge base it is authorized to use.

Routing patient inquiries through a supervisor agent supports high concurrency and extensibility. New departments or features can be added by introducing new collaborator agents and knowledge bases without redesigning the entire system. Because routing is handled centrally, changes in classification logic do not require updates across many independent supervisors.

Using RAG within each collaborator agent ensures that responses are grounded in department-approved information sources, which is critical in healthcare settings to reduce unsafe or incorrect guidance. This approach also improves performance because each retrieval scope is smaller and more relevant, supporting thousands of parallel interactions.

Option B introduces manual handoffs that do not scale. Option C relies on rule-based routing inside one general agent, which becomes brittle and difficult to govern as complexity grows. Option D mixes all departments into a single knowledge base and merges responses externally, increasing risk of incorrect domain answers and operational overhead.

Therefore, Option A best meets the scalability, correctness, and multi-agent onboarding requirements.

Option B is correct because the requirement is agentic, dynamic orchestration rather than a predefined workflow. AgentCore Memory can preserve conversational context, while AgentCore Gateway provides a unified MCP interface through which an agent can discover and invoke many tools and backend APIs. AWS documents support multiple Gateway targets and consolidated tool discovery, so adding or selecting enterprise capabilities does not require hard-coding every workflow path. AgentCore observability also exposes detailed traces covering processing steps, tool invocations, decision points, errors, and outputs, which supports compliance review and explainability. Option A is fundamentally static orchestration. Option C substitutes retrieval for API execution and still requires custom Lambda orchestration. Option D uses guardrails for a purpose they are not designed to perform; guardrails constrain content but are not a dynamic enterprise-tool orchestration engine. AWS Documentation

================

Option B is the correct solution because it directly addresses both throughput bottlenecks and latency requirements using native Amazon Bedrock performance optimization features that are designed for real-time, high-volume generative AI workloads.

Amazon Bedrock supports cross-Region inference profiles, which allow applications to transparently route inference requests across multiple AWS Regions. During peak usage periods, traffic is automatically distributed to Regions with available capacity, reducing throttling, request queuing, and timeout risks. This approach aligns with AWS guidance for building highly available, low-latency GenAI applications that must scale elastically across geographic boundaries.

Token batching further improves efficiency by combining multiple inference requests into a single model invocation where applicable. AWS Generative AI documentation highlights batching as a key optimization technique to reduce per-request overhead, improve throughput, and better utilize model capacity. This is especially effective for lightweight, low-latency models such as Claude 3 Haiku, which are designed for fast responses and high request volumes.

Option A does not meet the requirement because purchasing provisioned throughput in a single Region creates a regional bottleneck and does not address multi-Region availability or traffic spikes beyond reserved capacity. Retries increase load and latency rather than resolving the root cause.

Option C improves application-layer scaling but does not solve model-side throughput limits. Client-side round-robin routing lacks awareness of real-time model capacity and can still send traffic to saturated Regions.

Option D is unsuitable because batch inference with asynchronous retrieval is designed for offline or non-interactive workloads. It cannot meet a strict 2-second response time requirement for an interactive AI assistant.

Therefore, Option B provides the most effective and AWS-aligned solution to achieve low latency, global scalability, and high throughput during peak usage periods.