AIP-C01 AWS Certified Generative AI Developer - Professional Questions and Answers

Option A is correct because the problem is not total model runtime; the problem is perceived UI responsiveness. AWS documentation states that the Amazon Bedrock InvokeModelWithResponseStream API returns the model response as a stream. AWS also explains that streaming allows an application to access responses in chunks without waiting for the entire result. This lets the frontend display generated content as tokens or chunks arrive, improving the user experience even when the total inference duration remains unchanged.

This option also preserves the stated constraints. It does not require changing the foundation model, increasing token counts, provisioning extra compute, or reducing the model’s actual inference time. Instead, it changes how the response is delivered to the user. This is exactly the right pattern when users perceive slowness because the interface waits for the full completion before rendering anything. AWS examples for Bedrock streaming show applications processing response streams in real time, which supports interactive applications during high-concurrency periods.

The monitoring portion of option A is also aligned with AWS guidance. Amazon Bedrock runtime publishes CloudWatch metrics for invocation volume, latency, token consumption, error rates, and model invocation logging delivery. CloudWatch GenAI observability can track invocation count, token usage, and errors using built-in views. This supports audit and compliance visibility across Regions.

Option B adds caching of partially processed fraud explanations, which can be risky for regulated transaction-specific explanations and does not solve the UI issue for unique requests. Option C changes inputs by condensing them, which violates the constraint not to modify token counts or inputs. Option D focuses on Lambda concurrency and log retention, which does not address waiting for the full model response. Therefore, response streaming with CloudWatch and tracing is the correct solution.

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.

==========

This error occurs because SCPs override IAM permissions, and the SCP currently blocks Bedrock inference calls that resolve to eu-west-3, even though the company intends to use cross-Region inference (CRI) from eu-central-1.

Amazon Nova Pro is not hosted in eu-central-1, so when invoked, Amazon Bedrock transparently routes the request to a supporting Region (such as eu-west-3) through CRI inference profiles. However, SCPs that restrict Regions or specific Bedrock resources will block this routing unless explicitly allowed.

Option B is required because the SCP must explicitly allow the eu.amazon.nova-pro-v1:0 inference profile, which is the Bedrock abstraction that enables CRI while preserving data residency guarantees. Without this, Bedrock cannot legally route the request.

Option E is also required to allow EU-scoped inference profiles rather than individual Regions. This preserves precise governance while allowing Bedrock-managed CRI routing within the EU boundary, ensuring no data leaves Europe.

Option A violates least-privilege and does not override SCPs. Option C breaks data residency by enabling direct eu-west-3 access. Option D does not resolve the SCP denial.

Therefore, Options B and E are the only combination that resolves the error while preserving governance and EU-only data residency.

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 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 best meets the combined requirements of low latency, stability, and validated safety controls by using purpose-built Amazon Bedrock features designed for production GenAI operations. The company’s latency target of under 1 second and its observation of degradation during spikes strongly indicate capacity and throughput variability. Provisioned throughput for Amazon Bedrock is intended to deliver more predictable performance by reserving inference capacity for a chosen model, reducing throttling risk and stabilizing response times under load. This directly improves operational consistency across Regions where on-demand capacity can vary.

The requirement to “block unsafe or hallucinated recommendations” is most directly addressed by Amazon Bedrock Guardrails . Guardrails provide managed safety enforcement, including sensitive information controls and configurable content policies. Using semantic denial rules enables the application to prevent unsafe guidance such as dangerous brewing temperatures or other harmful procedural instructions, enforcing safety at the model boundary rather than relying on downstream filtering.

The remaining requirement is “99.5% output consistency for identical inputs.” While generative models can be probabilistic, production systems achieve practical consistency by controlling prompt versions, inputs, and policy behavior. Amazon Bedrock Prompt Management supports controlled prompt lifecycle practices, including versioning and approval workflows, which reduce unintended drift across deployments and Regions. By ensuring the same approved prompt templates and parameters are used consistently, the company can materially improve repeatability for the same structured inputs and retrieval context, which is essential in multi-stage prompt chains.

The other options are incomplete. B improves experimentation and observability but does not enforce safety controls or stabilize latency. C can improve performance, but it does not provide validated safety enforcement at inference time. D can help retrieval relevance, but it does not address unsafe outputs or inference stability. Therefore, A is the only option that simultaneously targets predictable latency, governance of prompt behavior, and strong safety controls within Amazon Bedrock.

Option A is the appropriate AWS-native approach because it keeps the authoritative documents in an Amazon S3 bucket located in the regulated Region while using managed Amazon Bedrock foundation-model infrastructure instead of maintaining containerized models. Amazon Bedrock supports private connectivity through AWS PrivateLink interface VPC endpoints. Applications inside the VPC can therefore reach Bedrock privately without requiring an internet gateway, NAT gateway, public IP address, or public internet path.

Amazon Bedrock Knowledge Bases can use Amazon S3 as a RAG data source. AWS requires the S3 bucket used by a knowledge base to be in the same Region as the Amazon Bedrock knowledge base. During ingestion, source content is converted into vector embeddings and stored in the configured vector store so similarity-based retrieval can locate relevant passages for augmentation. This creates an AWS-managed RAG architecture while preserving explicit regional placement of the underlying source documents.

Option B introduces AWS Outposts unnecessarily. More importantly, AWS documentation states that S3 on Outposts object data physically remains on the Outpost and “is not in an AWS Region.” That conflicts with a requirement written specifically as keeping data within a particular AWS Region and also introduces additional infrastructure and operational cost.

C fails because “a Region close to” the regulated Region does not meet a requirement for the specified Region. Encryption does not override residency requirements. D requires the company to deploy and operate foundation-model containers, directly contradicting the objective of minimizing model infrastructure and maintenance.

Therefore, regional S3 storage combined with privately accessed managed Amazon Bedrock services provides the closest match to the stated residency, RAG, cost, and operational requirements.

==========

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

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

Amazon Bedrock Flows—referred to as “Amazon Bedrock Prompt Flows” in the question—is the lowest-complexity solution because the workflow consists primarily of generative-AI processing stages and one straightforward conditional branch. Bedrock Flows provides a managed visual workflow environment in which nodes represent steps that invoke Amazon Bedrock or related resources. This eliminates the need to build a separate orchestration service merely to connect multiple model interactions.

A prompt node defines a prompt, receives values as input variables, invokes the configured model, and produces the model completion as its output. Therefore, separate prompt nodes can represent document analysis, structured data extraction, fraud analysis, and recommendation generation.

A condition node provides deterministic branching. AWS documents relational operators including > , > = , < , < = , == , and != , and the node can direct its input to different downstream nodes based on the configured condition. The application can therefore test whether the claim amount is greater than $10,000. Claims above that threshold are routed through the fraud-analysis node before reaching the recommendation stage; other claims can go directly to recommendation generation.

A Step Functions implementation in A would work technically and is appropriate when orchestration spans a broad set of distributed AWS services. However, for a Bedrock-centric GenAI workflow, it adds another orchestration layer and therefore is not the least-complex solution. B unnecessarily introduces autonomous agent reasoning for deterministic sequential processing. D requires custom Lambda invocation code, error handling, routing logic, deployment, and maintenance.

Because the required workflow maps directly to Bedrock ' s managed prompt and condition nodes, C minimizes infrastructure and custom orchestration while preserving deterministic control over the $10,000 fraud-analysis requirement.

==========

Option C is the technically supportable choice under current AWS documentation. Amazon Bedrock prompt caching does not provide semantic caching for differently worded or cross-language questions; it reuses cached prompt content through matching prompt prefixes or defined cache checkpoints. Therefore, Option B ' s assumption that Bedrock prompt caching automatically recognizes semantically equivalent multilingual queries is incorrect, even though many supported models provide a 5-minute prompt-cache TTL. ElastiCache for Redis OSS supports application-controlled cache entries and TTLs, while Global Datastore provides managed cross-Region replication and low-latency geographic reads. A multilingual semantic fingerprint can normalize equivalent requests to a shared cache key before inference. Options A and D similarly lack built-in semantic equivalence handling, while DAX is specifically a DynamoDB caching layer. Thus, C is the viable architecture among the listed choices.

AWS Glue Data Quality is the correct core service because it directly addresses the distinctive requirements that separate this question from ordinary ETL validation. AWS describes Glue Data Quality as a managed, serverless capability for measuring and monitoring dataset quality. It uses Data Quality Definition Language (DQDL) for rules, can automatically analyze datasets and recommend rules, generates a Data Quality score after rule evaluation, supports machine-learning-based anomaly detection, and integrates with Amazon CloudWatch.

Completeness can be evaluated through DQDL rules such as IsComplete . AWS rule recommendations can automatically generate completeness, uniqueness, value-domain, and column-length rules based on observed data. Anomaly detection learns historical statistics and identifies values outside predicted bounds; AWS can additionally recommend future data-quality rules based on detected anomalies.

Glue Data Quality can publish pass/fail metrics into the Glue Data Quality CloudWatch namespace, making the results usable for monitoring, alarms, and regulatory-quality dashboards.

There is one important technical issue in the wording of option B: AWS Glue Data Quality itself does not remove PII. The complete AWS Glue implementation would pair Glue Data Quality with the AWS Glue Detect PII transform. AWS documents that this transform can identify sensitive entities and then mask, redact, hash, or otherwise process identified PII. Therefore, B is clearly the intended and strongest listed answer, but a production implementation should add the Glue PII transform to fully satisfy the question ' s PII-removal requirement.

A requires extensive custom validation. C does not provide the requested automated rule recommendations and managed quality-score capabilities. D handles PII detection but lacks automated DQDL recommendations, ML anomaly detection, and native data-quality scoring.

==========

Amazon Bedrock Prompt Management is the purpose-built service for prompt lifecycle management. it provides native version control , allowing developers to test and compare variants side-by-side. Use of parameterized variables ensures that a single prompt structure can be consistently reused across different sub-applications (assistants vs. graders) while still being tailored to the specific context. To " integrate into the main application " with minimal overhead, Amazon Bedrock Flows provide a managed orchestration layer. Flows allow developers to link managed prompts with AWS services (like knowledge bases or Lambda functions) without writing complex state-machine logic in Step Functions (Option B). This combination ensures consistent, auditable, and easily deployable prompt assets across the university ' s diverse use cases.

Option A addresses all three governance layers. Model cards provide documented information about a model’s capabilities, limitations, supported modalities, context constraints, and other characteristics that teams can review before deployment. Amazon Bedrock Guardrails can enforce application-specific safeguards on user inputs and model responses, and IAM policies can require a specific guardrail for inference calls, helping prevent uncontrolled model use. Storing generated outputs in Amazon S3 provides durable audit storage, while S3 Object Lock uses a write-once-read-many model that can prevent retained objects from being overwritten or deleted and is specifically intended to support regulatory retention requirements. Option B lacks formal model-limit documentation and relies on custom postprocessing. Option C does not protect stored audit evidence against tampering. Option D relies primarily on prompts and ordinary database storage rather than enforceable safety and immutable-retention controls. AWS Documentation

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

Option B best meets all requirements because it combines AWS-recommended resiliency patterns for transient failures with streaming-aware handling and adaptive protection against cascading retries during peak load. When timeouts and throttling occur, naïve retries can amplify traffic and worsen outages. Exponential backoff with jitter is the standard AWS best practice because it spreads retry attempts over time, reduces synchronized retry storms, and lowers the probability of repeatedly colliding with service limits.

The requirement also states the strategy must “adapt to changing service availability” and “prevent overwhelming Amazon Bedrock.” A circuit breaker pattern directly addresses this by temporarily stopping or reducing retries when failure rates exceed a threshold, allowing the system to degrade gracefully instead of continually hammering the service. This is a key mechanism to prevent cascading failures during throttling events.

Because the application uses response streaming and experiences broken chunks, the retry strategy must be streaming-aware. A streaming response handler that detects chunk delivery timeouts and buffers already received chunks prevents the user from losing progress when a connection drops. Resuming from the last successfully received chunk minimizes redundant generation and reduces additional load on the model compared with restarting the entire stream. This supports better user experience and better service efficiency during intermittent failures.

Token-aware request handling is supported in this architecture because the application can apply token budgeting before invoking the model (for example, trimming or summarizing excessive context) while still preserving streaming output behavior. Option B provides the correct backbone for this by focusing on adaptive control and robust streaming recovery.

Option A is too simplistic and risks retry storms. Option C combines conflicting elements (global token limit, cached completions for streaming) and includes impractical “request only missing chunks” behavior that is not a reliable property of streamed generative output. Option D includes useful ideas (load shedding) but relies on static caps and does not provide as strong adaptive retry control as circuit breaking.

Therefore, Option B is the most correct and operationally safe strategy for peak-load Bedrock streaming workloads.

Option B best meets the requirements because it provides systematic diagnosis, measurable comparison, and historical traceability of prompt performance. By placing prompts under version control and testing them against complex clinical documents, the company can consistently reproduce issues, track regressions, and compare prompt behavior using quantifiable metrics such as factual accuracy, completeness, and consistency. Automated testing ensures scalability and repeatability, while version history preserves prompt evolution over time.

Option A lacks objective metrics and does not address complex documents. Option C focuses on live traffic experimentation but does not inherently diagnose prompt inconsistencies or preserve detailed historical evaluations. Option D adds medical entity analysis but introduces unnecessary service coupling and does not provide robust prompt version history or automated comparative benchmarking. Therefore, Option B is the most complete and disciplined solution.

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 A best satisfies the requirement to capture multi-hop, highly interconnected relationships with minimal operational overhead. Traditional vector similarity search excels at finding semantically similar text but is not optimized for reasoning over explicit entity-to-entity relationships, especially when analysts need indirect, multi-hop connections (for example, fund → holding → issuer → sector → regulation). Graph-based retrieval is designed specifically for these kinds of relationship traversals.

GraphRAG combines retrieval-augmented generation with graph-aware context selection. By representing entities and their relationships in a graph store, the system can traverse multiple hops to assemble a holistic set of relevant facts. This improves completeness and reduces the chance that the model misses indirect relationships that are essential for accurate investment guidance.

Amazon Neptune Analytics provides a managed graph analytics environment capable of efficiently traversing and analyzing complex relationship networks. When integrated with Amazon Bedrock Knowledge Bases, it reduces custom engineering by providing managed ingestion, retrieval, and orchestration patterns suitable for GenAI applications. This lowers operational overhead compared to building and maintaining custom multi-stage retrieval logic.

Meeting the sub-3-second requirement is also more feasible with a graph-optimized engine because multi-hop traversals can be executed efficiently compared to chaining multiple vector searches and joining results in an application layer. The managed nature of Knowledge Bases and Neptune Analytics reduces maintenance, scaling, and operational burden while enabling strong performance.

Option B and C require extensive custom logic and orchestration, increasing complexity and latency. Option D is not designed for graph-style multi-hop exploration and would require significant custom indexing and retrieval logic.

Therefore, Option A is the most AWS-aligned and operationally efficient approach for multi-hop relationship-aware RAG with strong performance.

Option A best matches the requirement because managed knowledge-base retrieval provides source attribution and lets access be governed through IAM, avoiding custom result-merging logic. Amazon Bedrock RetrieveAndGenerate returns citations tied to retrieved source references. One implementation detail matters under current AWS documentation: structured Knowledge Bases use Amazon Redshift as the query engine; Amazon Redshift federated queries can access Amazon RDS and Aurora operational databases. Therefore, the RDS-backed knowledge-base leg should be implemented through that supported structured-data path rather than interpreted as a direct native RDS vector connector. Options B, C, and D introduce extra agents or Lambda-based retrieval/merging, which increases development effort and makes unified citation handling more custom. A keeps both information sources behind managed retrieval and preserves provenance while IAM controls access. AWS Documentation

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

Option A best meets the requirements because it applies an AWS-aligned multi-agent pattern that cleanly separates responsibilities: a supervisor agent performs intent classification and orchestration, while specialized collaborator agents handle domain-specific tasks using the right knowledge sources. This structure is well suited for healthcare workflows where clinical questions, scheduling, and insurance processes require different policies, terminology, and data access boundaries.

The requirement for appropriate domain-specific responses is addressed by routing each user query to a department-focused collaborator agent that is grounded with its own department-spe cific knowledge base . Using Retrieval Augmented Generation with the correct knowledge base improves factual alignment and reduces cross-department leakage (for example, avoiding claims content in a clinical answer). It also supports better prompt grounding and more consistent tone and constraints per department.

The requirement to isolate data maps to using separate knowledge bases per agent and enforcing access through IAM controls , ensuring that each agent can retrieve only from the authorized datasets. This is important for minimizing unintended exposure of sensitive or irrelevant departmental data and supports governance and compliance needs.

For scalability and thousands of parallel interactions , this architecture minimizes contention and bottlenecks. Each collaborator agent can scale independently because requests are distributed across multiple agents and multiple retrieval backends. Operationally, onboarding new features is also simpler: the company can add a new collaborator agent (for example, “billing disputes” or “pharmacy refills”) with its own knowledge base and policies without redesigning the entire assistant.

Option B introduces unnecessary complexity with multiple supervisors and manual handoffs. Option C overloads a single agent with broad instructions and rule-based routing, which increases prompt complexity and reduces maintainability as features grow. Option D creates high operational complexity and risks inconsistent outputs when merging responses from parallel supervisors, and it weakens data isolation by using a shared knowledge base across agents.

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 provides the strongest modular architecture for an integration-heavy agentic workload. Strands supports hierarchical agents-as-tools patterns in which a coordinating agent delegates specialized work to independent agents, allowing integrations and domain logic to be added or modified without creating one increasingly complex agent. AgentCore Gateway exposes tools through standardized MCP interfaces so agents can discover and invoke internal or external capabilities through a unified connectivity layer. AWS Step Functions supports callback tasks that pause execution until a human or external process returns a task token, making it suitable for the required human approval checkpoint. Option A requires custom model hosting and approval UI development. Option C is primarily a batch data-processing architecture rather than real-time agent orchestration. Option D can function initially, but a single agent becomes less modular and maintainable as the number of tools and business domains expands. Strands Agents

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

For a 3-week proof-of-concept in a regulated field like healthcare, Retrieval Augmented Generation (RAG) is more efficient and safer than fine-tuning. RAG allows the use of anonymized patient records without risking the leak of sensitive data into the model ' s permanent memory. To evaluate accuracy quantitatively and rapidly, the " LLM-as-a-judge " pattern is recommended. Using a strong judge model to score the outputs of multiple candidate FMs provides objective metrics (e.g., factual alignment, completeness) that manual qualitative feedback (Option C) cannot scale to provide within the timeline. Fine-tuning (Option B) typically takes longer than 3 weeks to properly data-prep and validate for clinical accuracy.

Option D is the best choice because it delivers true semantic search with the smallest operational footprint by combining a fully managed embedding service with an automatically scaling vector-capable database. The university’s requirement is explicitly semantic: the metadata has no keywords, and the system must match abstracts based on similarity of meaning. This is a direct fit for an embeddings-based approach, where each abstract is converted into a vector representation and searched using vector similarity. Amazon Titan Embeddings in Amazon Bedrock provides a managed way to generate these vectors without hosting or maintaining an ML model, eliminating the operational work of model provisioning, patching, scaling, and lifecycle management.

For storage and retrieval, Amazon Aurora PostgreSQL Serverless with the pgvector extension supports vector storage and similarity search while minimizing infrastructure operations. Aurora Serverless reduces capacity planning and scaling tasks because it can automatically adjust to changes in workload, which is valuable for a university search application with variable usage patterns. With fewer than 1 million files, a PostgreSQL-based vector store is commonly operationally simpler than running a dedicated search cluster, while still meeting the requirement to query using both text-derived similarity and associated metadata filters stored alongside the vectors.

Option A can also enable vector search, but operating an OpenSearch domain typically introduces additional concerns such as domain sizing, shard strategy, cluster scaling, and performance tuning for k-NN workloads. Option C increases operational overhead the most because it requires deploying and operating a sentence-transformer model endpoint in SageMaker AI, including scaling, monitoring, and model management. Option B does not meet the semantic similarity requirement reliably because topic extraction is not equivalent to embedding-based semantic matching, especially when the metadata lacks keywords and the system must compare abstracts by meaning.

Therefore, D best satisfies semantic search needs with the least operational overhead.

Option D is the correct solution because the Amazon Bedrock Converse API is purpose-built for multi-turn conversational interactions and is designed to work consistently across SDKs and compute environments. The Converse API standardizes how messages, roles, and context are represented, which ensures consistent behavior whether the application is running in AWS Lambda with Python or in Amazon EKS with JavaScript.

By passing previous messages in the messages array, the application explicitly maintains conversational context across turns without relying on external state stores. This approach is recommended by AWS for conversational GenAI workflows because it avoids state synchronization complexity and ensures deterministic model behavior across environments.

Using IAM roles for authentication provides a single, consistent security model for both Lambda and EKS. IAM roles integrate natively with AWS SDKs, eliminating the need for custom authentication logic or environment-specific credentials. This aligns with AWS best practices for least privilege and simplifies governance.

Option A introduces inconsistent authentication and custom formatting logic, increasing complexity. Option B unnecessarily introduces ElastiCache for state management, which is not required when using the Converse API correctly. Option C stores state in process memory, which is unsafe and unreliable for serverless and containerized workloads.

Therefore, Option D best satisfies the requirements for conversational consistency, multi-environment support, shared model usage, and consistent authentication with minimal operational overhead.

Option B meets the requirement to combine scalable automated evaluation with targeted human oversight using managed AWS GenAI capabilities. Amazon Bedrock evaluations enable systematic, repeatable quality assessment across large volumes of interactions. Using an LLM-as-a-judge approach with a strong evaluator model such as Anthropic Claude Sonnet allows the company to automatically score outputs for dimensions like factual accuracy, conversational appropriateness, and policy alignment. This directly supports “automated quality evaluations at scale” without building custom scoring models.

However, financial recommendations add higher risk because regulatory compliance requires additional enforcement beyond general quality scoring. Amazon Bedrock guardrails provide a dedicated policy enforcement layer that can block or intervene when responses violate compliance constraints. Guardrails are particularly important for preventing disallowed financial guidance patterns and ensuring consistent behavior across deployments.

The requirement also calls for “targeted human reviews of critical interactions.” Amazon Augmented AI (A2I) is a managed human review service that supports routing specific items to human reviewers based on rules or confidence thresholds. In this design, the system can automatically send only high-risk or policy-flagged interactions to qualified financial experts for review, keeping human effort focused where it matters most while maintaining scale.

Option A is not scalable because it requires manual review of all responses. Option C relies on static rules and end-user feedback, which is insufficient for regulatory compliance and factual accuracy assurance. Option D provides monitoring but not structured quality evaluation or policy enforcement.

Therefore, Option B provides the most complete, AWS-aligned solution for scalable evaluation plus human oversight in a regulated financial context.

Option C combines AWS-native capabilities for the three required monitoring dimensions with minimal custom code. Amazon Bedrock model invocation logging can capture request and response information and deliver invocation data to Amazon S3 or CloudWatch Logs. Bedrock Guardrails contextual grounding checks are specifically designed to detect and filter responses that are not sufficiently grounded in supplied source information; AWS explicitly lists summarization as a supported use case. CloudWatch provides token-usage observability, while anomaly-detection mechanisms can establish adaptive baselines instead of requiring manually maintained static thresholds. This enables earlier detection of unusual consumption and potential cost anomalies. Option A requires custom Athena-based hallucination analysis. Option B relies on evaluation jobs and custom Lambda processing rather than near-real-time managed controls. Option D combines services that do not directly provide the required Bedrock-specific hallucination monitoring. AWS Documentation

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

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 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 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.

Option C provides the strongest combination of residency controls, data classification, retention protection, and auditability. Keeping source data in Region-specific S3 buckets and processing it according to geographic origin establishes a clear architectural boundary around data belonging to Europe, North America, and Asia. Amazon S3 Object Lock provides write-once-read-many protection and can prevent protected object versions from being overwritten or deleted, which is particularly valuable for regulated records and audit evidence. Compliance mode provides especially strong retention enforcement.

Amazon Macie provides the required classification capability. Macie performs automated sensitive-data discovery against Amazon S3 and produces classification results describing whether analyzed objects contain sensitive information. It also calculates sensitivity scores and labels for monitored S3 buckets, supporting governance and compliance workflows.

AWS CloudTrail supplies auditable records of AWS API activity. When CloudTrail log-file integrity validation is enabled, CloudTrail uses SHA-256 hashing and digital signatures so modifications or deletions of delivered logs can be detected. Combined with protected S3 storage, this provides a strong evidentiary audit trail.

AWS now also supports geographic cross-Region inference profiles for Amazon Bedrock that keep inference processing within defined geographic boundaries such as the US, EU, and APAC. However, option A merely states “cross-Region inference” and lacks the required classification service while relying on manual compliance tracking. Geographic residency must be explicitly controlled rather than assumed.

B does not provide a data-classification mechanism. D creates significantly greater administrative burden and depends on custom monitoring and manual reporting. Therefore, C best satisfies the combined residency, classification, audit, and compliance requirements.

==========

Option C directly addresses both retrieval relevance and performance scalability. Fixed token chunking breaks semantic continuity in legal texts, causing incomplete context retrieval and degraded response quality. By switching to semantic chunking—based on legal arguments, clauses, or sections—the application preserves contextual integrity, improving retrieval accuracy and reducing hallucinations.

Regenerating embeddings aligned with the new chunk structure also improves vector search efficiency, reducing unnecessary comparisons and helping control latency as the dataset scales.

Option A increases cost and latency without fixing the core issue. Option B removes dynamic reasoning, which defeats the purpose of a legal RAG system. Option D discards vector semantics entirely and is unsuitable for nuanced legal research. Therefore, Option C is the correct and scalable solution.

The first option labeled B is the correct solution because it fully satisfies both private connectivity and fine-grained cross-account data governance requirements using AWS-native services.

Creating interface VPC endpoints for Amazon Bedrock runtimes ensures that all inference calls remain on the AWS private network and never traverse the public internet. Running AWS Lambda functions in private subnets enforces network isolation, and using IAM conditions that restrict access to specific VPC endpoints and roles prevents unauthorized inference calls.

For the governed data lake, AWS Lake Formation LF-tag–based access control is the recommended AWS mechanism for enforcing cross-account, column-level permissions. LF-tags allow the company to define data access policies once and apply them consistently across accounts, databases, tables, and even individual columns. This is required for sensitive customer records and is not achievable with S3 bucket policies or IAM alone.

The second option labeled B uses a NAT gateway, which violates the private connectivity requirement. Option C uses public Bedrock endpoints and only database-level grants, which are insufficient. Option D relies on IAM path-based policies, which cannot enforce column-level access and introduces public fallback paths.

Therefore, the first option labeled B is the only solution that meets all networking, security, and data governance requirements.

Option D is correct because it separates the United States and European workloads and uses the correct Amazon Bedrock inference pattern for European data residency. AWS documentation states that geographic cross-Region inference keeps data processing within specified geographic boundaries such as US, EU, and APAC, and is intended for organizations with data residency and compliance requirements. The regional availability documentation further explains that a geographic profile routes within a defined geography and that prompts and outputs may move within that geography but not outside it. This directly satisfies the requirement that European customer data must be processed only in Europe-based AWS Regions.

Provisioned throughput is appropriate for the US workload because the system has a demanding and predictable performance target: 1,000 transactions per second with sub-500 ms response times. Provisioned capacity is generally used when an application needs reserved throughput and consistent performance rather than depending only on on-demand capacity. Automatic scaling based on utilization metrics supports changing demand while maintaining performance. EventBridge and Lambda can coordinate failover workflows or routing updates when connectivity disruptions occur, supporting high availability across Regions.

Option A is not the best answer because it includes a global inference profile. AWS documentation states that global cross-Region inference can route requests across supported commercial AWS Regions worldwide and has no geographic data-residency restriction. AWS explicitly notes that organizations with data residency or compliance requirements should assess whether global cross-Region inference fits their framework because requests may be processed in other supported commercial Regions. That conflicts with the European data processing requirement.

Option B uses separate Regional provisioned throughput, but it relies on a custom Application Load Balancer approach for Bedrock routing and does not use the Bedrock geographic inference profile that is designed for EU data residency. Option C uses response streaming, which improves perceived response latency for text generation but does not guarantee sub-500 ms fraud-processing latency or European-only processing. Therefore, option D is the most complete AWS-aligned solution.

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.

Options A and C provide the most AWS-native governance approach. Amazon Bedrock Prompt Management supports reusable parameterized prompts and explicit prompt versions, allowing teams to preserve approved configurations and deploy stable versions instead of uncontrolled working drafts. An organizational approval gate can therefore be applied before a new prompt version becomes production-approved. AWS CloudTrail records Amazon Bedrock API activity, including the identity making the request, the operation performed, time, source information, and other event details, providing the required audit trail for Bedrock resource modifications and use. CloudWatch can complement this with operational usage monitoring and reporting. Option B recreates prompt governance through S3, Lambda, and SNS. Option D uses general-purpose configuration storage rather than Bedrock-native prompt management. Option E adds substantial custom state and filtering logic without providing the same managed prompt-version lifecycle. AWS Documentation

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

Option B is the optimal solution because it maximizes similarity search accuracy and performance for a small, proprietary dataset while maintaining low operational complexity. Amazon MemoryDB is a fully managed, in-memory database that provides microsecond-level latency, making it ideal for real-time RAG workloads that require fast vector similarity searches.

For small datasets with low index counts, the Hierarchical Navigable Small World (HNSW) algorithm is recommended by AWS for its high recall and accuracy. Unlike approximate methods optimized for massive datasets, HNSW excels at returning the most semantically relevant vectors with minimal loss of precision, which directly improves the quality of responses generated by the Amazon Bedrock foundation model.

Vertical scaling in MemoryDB is sufficient for this use case because the dataset size is limited. Scaling up instance size provides increased memory and compute capacity without the complexity of managing distributed indexes or sharding strategies. This simplifies operations while maintaining predictable performance.

Option A’s Flat algorithm is computationally expensive and inefficient at scale, even for moderate query volumes. Option C introduces higher latency and operational overhead by using a relational database not optimized for in-memory vector search. Option D is unsuitable because Amazon DocumentDB is not designed for high-performance vector similarity workloads and introduces unnecessary replica management complexity.

Therefore, Option B best meets the requirements for accuracy, performance, and efficient integration with an Amazon Bedrock–based RAG application.

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 B best satisfies the requirements because it directly applies Retrieval Augmented Generation principles using managed Amazon Bedrock Knowledge Bases, which are designed to handle large, complex documents while preserving contextual relationships. Financial reports often in terleave tables with explanatory narrative, and accurate analysis depends on keeping those elements logically connected. By segmenting documents based on their structural layout —for example, sections, subsections, tables, and surrounding commentary—the knowledge base can retrieve semantically relevant chunks that maintain this relationship during inference.

Amazon Bedrock Knowledge Bases support contextual chunking strategies that go beyond simple fixed-size segmentation. This is critical for financial documents, where a metric in a table may be explained in adjacent paragraphs or footnotes. Context-aware chunking ensures that retrieved content includes both the numeric data and its interpretation, enabling the foundation model to generate accurate, grounded responses. Including citations further improves analyst trust and auditability by allowing users to trace answers back to specific source sections, which is a common requirement in financial environments.

Scalability is another key requirement. Knowledge Bases manage embedding generation, indexing, and retrieval orchestration as a managed service, which allows the solution to scale across large document collections without requiring custom infrastructure or model hosting. This approach also supports efficient updates as new quarterly reports are added, ensuring the retrieval layer remains current.

Option A does not scale well because processing entire 5–100 page documents in a single prompt increases token usage, latency, and cost while risking context truncation. Option C relies on fixed-size chunking triggered at query time, which often breaks semantic relationships in structured financial content. Option D introduces unnecessary architectural complexity by splitting structured and unstructured data into separate applications, increasing operational overhead without providing better contextual retrieval than a unified RAG approach.

Option B is correct under current AWS capabilities. Amazon Bedrock Custom Model Import accepts specified supported model architectures; it is not a mechanism for importing an arbitrary proprietary architecture. SageMaker inference supports bringing a custom container, making it appropriate when custom model architecture or inference code must be preserved. Amazon Bedrock Guardrails can be applied independently of Bedrock-hosted foundation models through the ApplyGuardrail API, including to output from third-party or externally hosted models. Contextual grounding checks calculate grounding and relevance scores and can block responses below configured thresholds, directly addressing hallucination detection. Options A and C incorrectly use ordinary content filters for grounding and relevance, which are separate contextual-grounding capabilities. Option D incorrectly assigns contextual-grounding functionality to AWS HealthScribe rather than Amazon Bedrock Guardrails. AWS Documentation

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

AWS Glue Data Catalog and its associated crawlers are the standard AWS tools for automatic discovery and centralized cataloging of datasets. For the generated summaries, storing them in Amazon S3 allows the use of object tags for metadata (like source IDs), making them easily queryable. The critical requirement for " tamper-evident, immutable audit logs " is met by enabling Bedrock model invocation logging to an S3 bucket protected by S3 Object Lock (compliance mode). To further guarantee that logs have not been altered, AWS CloudTrail log file integrity validation uses cryptographic hashes to provide non-repudiation and a verifiable audit trail. This combination covers data management, metadata attribution, and high-standard security compliance.

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 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 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.

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 Amazon Q Developer customizations are designed to incorporate organization-approved knowledge and coding guidance without requiring per-project changes. A customization can point Amazon Q Developer to curated internal sources such as approved libraries, coding standards, architectural patterns, and proprietary techniques. This allows the assistant’s suggestions to align with company policies and preferred implementations consistently across teams and repositories.

The key requirement is that the company does not want to make project-level modifications. Options A, B, and C all require adding files or repositories into the project workspace, which directly violates this constraint. They also rely on developer behavior to “use workspace context,” which is harder to enforce and can lead to inconsistent adherence to standards.

With a customization, the organization centrally manages and updates approved resources. This reduces operational overhead because updates to libraries, patterns, or guidelines propagate automatically to developers using the customization, without requiring changes to each project. This is especially valuable for a new team, where consistent enforcement of approved practices is important to reduce compliance risk, security issues, and inconsistent code style.

Additionally, customizations support governance by allowing the company to standardize how Amazon Q Developer responds, ensuring that suggestions reflect approved internal content rather than generic public patterns.

Therefore, Option D best satisfies the requirement for centralized enforcement of approved resources with minimal ongoing management and no project-level modifications.