Amazon Web Services AIP-C01 - AWS Certified Generative AI Developer - Professional

Option A is the correct solution because it uses native Amazon S3 metadata mechanisms to create a consistent, queryable, and model-friendly metadata framework with minimal complexity. S3 system metadata automatically records object creation and modification timestamps, providing reliable and consistent temporal context without additional processing.

Custom user-defined metadata is the appropriate mechanism for storing structured attributes such as author information. These key-value pairs are stored directly with the object, remain consistent across uploads, and can be accessed programmatically by downstream indexing or retrieval systems used by GenAI applications.

S3 object tags are ideal for domain classification because they are designed for lightweight categorization, filtering, and access control. Tags can be standardized across the organization to ensure consistent research domain labeling and can be consumed by search indexes or knowledge base ingestion pipelines without requiring access to the full document body.

Together, system metadata, user-defined metadata, and object tags provide a clean separation of concerns: timestamps for temporal context, metadata for authorship, and tags for classification. This structure allows foundation models to reason about document context (such as recency, domain relevance, and authorship) based on metadata alone, improving retrieval precision and reducing unnecessary token usage.

Options B, C, and D misuse features like Object Lock, access points, Storage Lens, or event notifications for purposes they were not designed for, adding complexity without improving metadata quality or model understanding.

Therefore, Option A best satisfies the metadata consistency, context enrichment, and low-overhead requirements for GenAI-driven document analysis.

The correct answers are A and D because they directly reduce time-to-first-token and stabilize p95 latency for interactive, real-time chat workloads hosted on Amazon SageMaker AI real-time endpoints.

Option D addresses the biggest driver of uneven latency: cold starts and scale-to-zero behavior. By setting the minimum number of instances to greater than 0, the endpoint always has warm capacity and loaded runtime resources, eliminating the first-request penalty that causes users to wait multiple seconds. Enabling response streaming improves perceived latency by returning the first tokens as soon as they are generated rather than waiting for the complete response. This directly targets the abandonment problem described (users leaving after waiting for the first token).

Option A further improves p95 latency and throughput by removing model loading overhead during inference and improving GPU utilization. Preloading model weights during container startup ensures the model is ready before traffic arrives and avoids unpredictable on-demand weight loading. Dynamic batching increases efficiency by grouping compatible requests into a single inference pass, reducing per-request overhead and improving GPU saturation. When tuned properly for interactive workloads, batching can reduce tail latency while preserving responsiveness by enforcing small batch windows.

Option B makes latency worse because setting minimum instances to 0 and lazily loading weights guarantees cold-start delays and unpredictable first-token performance. Option C similarly increases cold-start behavior through lazy loading and offers no batching benefits. Option E is designed for non-interactive workloads and introduces queueing and storage latency, which conflicts with the 800 ms p95 requirement for interactive chat.

Therefore, A and D are the best combination to achieve consistently low p95 latency and fast first-token streaming for a SageMaker-hosted chat assistant.

Option C is the correct solution because it provides near real-time monitoring, hallucination detection, and cost anomaly awareness using built-in Amazon Bedrock and Amazon CloudWatch capabilities, with minimal custom development.

By configuring Amazon Bedrock invocation logging with text output logging, the company captures detailed prompt and response data for auditing and analysis without building custom logging pipelines. This data is stored in Amazon S3, providing durable storage for compliance and retrospective investigation.

Using Amazon Bedrock guardrails with contextual grounding checks allows the application to automatically detect hallucinations by verifying whether generated summaries are grounded in the provided clinical documents. This is the AWS-recommended approach for hallucination detection in RAG and summarization workloads and avoids the need to maintain custom evaluation models or pipelines.

Creating Amazon CloudWatch anomaly detection alarms for InputTokenCount and OutputTokenCount metrics enables automatic detection of abnormal token usage patterns that often correlate with runaway prompts, inefficient summarization, or prompt injection attempts. Anomaly detection adapts dynamically to usage trends, making it more effective than static thresholds for early cost warnings.

Option A introduces batch analytics with Glue and Athena, which is not near real time and increases operational overhead. Option B requires managing evaluation jobs and Lambda-based notification logic. Option D focuses on infrastructure-level monitoring and offline dashboards rather than near real-time GenAI quality and cost signals.

Therefore, Option C best meets the requirements with the least operational effort and maintenance overhead.