Question 280

At a university research portal, engineers need to deploy a large self-hosted model with significant GPU memory needs. The organization needs repeatable automated validation. The team wants a solution that can be validated through targeted test cases. What is the most appropriate production approach?

  1. Treat model deployment as a CPU-only static web hosting task
  2. Assume endpoint throughput is independent of generated token length
  3. Run a large GPU workload in an API Gateway mapping template
  4. Size accelerator memory, batching, concurrency and model-loading strategy explicitly

Explanation

Correct Answer

D:

Self-hosted LLM serving must account for memory, accelerator use and token generation dynamics. In a university research portal, the engineering team should verify this design against its stated objective to deploy a large self-hosted model with significant GPU memory needs, including the requirement that the organization needs repeatable automated validation.

Incorrect Answers

A:

The suggested action—treat model deployment as a CPU-only static web hosting task—does not establish the required control for a university research portal to deploy a large self-hosted model with significant GPU memory needs. It replaces a verifiable engineering control with an untested shortcut in the a university research portal deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

B:

The suggested action—assume endpoint throughput is independent of generated token length—does not establish the required control for a university research portal to deploy a large self-hosted model with significant GPU memory needs. The change is applied outside the component responsible for the requirement identified by a university research portal. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—run a large GPU workload in an API Gateway mapping template—does not establish the required control for a university research portal to deploy a large self-hosted model with significant GPU memory needs. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a university research portal. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 281

At a nonprofit grant portal, engineers need to allow an agent to read account status without permitting account deletion. The service must continue to support multiple independent tenants. The team wants a solution that can be validated through targeted test cases. What should be implemented before scaling this workflow?

  1. Rely on the system prompt as the sole permission boundary
  2. Give the agent direct unrestricted database administrator access
  3. Expose a constrained read-only tool and enforce least-privilege authorization
  4. Grant wildcard permissions so the agent can discover capabilities

Explanation

Correct Answer

C:

Tool authorization must be enforced by the execution identity, not only model instructions. In a nonprofit grant portal, the engineering team should verify this design against its stated objective to allow an agent to read account status without permitting account deletion, including the requirement that the service must continue to support multiple independent tenants.

Incorrect Answers

A:

The suggested action—rely on the system prompt as the sole permission boundary—does not establish the required control for a nonprofit grant portal to allow an agent to read account status without permitting account deletion. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a nonprofit grant portal. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

B:

The suggested action—give the agent direct unrestricted database administrator access—does not establish the required control for a nonprofit grant portal to allow an agent to read account status without permitting account deletion. It replaces a verifiable engineering control with an untested shortcut in the a nonprofit grant portal deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—grant wildcard permissions so the agent can discover capabilities—does not establish the required control for a nonprofit grant portal to allow an agent to read account status without permitting account deletion. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a nonprofit grant portal. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 282

At a retail bank, engineers need to maintain inference when the primary model experiences throttling. The next release must be evaluated against the previous baseline. The initial assessment isolates this issue from unrelated infrastructure changes. What should the lead engineer recommend?

  1. Disable all request timeouts
  2. Configure bounded retry with jitter and an eligible fallback model
  3. Retry without backoff until requests succeed
  4. Increase output token limits during throttling

Explanation

Correct Answer

B:

Bounded recovery avoids synchronized retry storms and enables graceful service degradation. In a retail bank, the engineering team should verify this design against its stated objective to maintain inference when the primary model experiences throttling, including the requirement that the next release must be evaluated against the previous baseline.

Incorrect Answers

A:

The suggested action—disable all request timeouts—does not establish the required control for a retail bank to maintain inference when the primary model experiences throttling. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a retail bank. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—retry without backoff until requests succeed—does not establish the required control for a retail bank to maintain inference when the primary model experiences throttling. The change is applied outside the component responsible for the requirement identified by a retail bank. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—increase output token limits during throttling—does not establish the required control for a retail bank to maintain inference when the primary model experiences throttling. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a retail bank. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 283

At a telecom support operation, engineers need to document model limitations and acceptable use before regulated deployment. The compliance team requires a traceable decision record. The initial assessment isolates this issue from unrelated infrastructure changes. What is the most appropriate production approach?

  1. Treat a model vendor marketing page as the complete risk assessment
  2. Skip governance if output is fluent
  3. Ask end users to discover limitations only after incidents
  4. Maintain model cards, risk assessments and accountable approvals

Explanation

Correct Answer

D:

Governance artifacts define limitations, ownership and acceptance criteria. In a telecom support operation, the engineering team should verify this design against its stated objective to document model limitations and acceptable use before regulated deployment, including the requirement that the compliance team requires a traceable decision record.

Incorrect Answers

A:

The suggested action—treat a model vendor marketing page as the complete risk assessment—does not establish the required control for a telecom support operation to document model limitations and acceptable use before regulated deployment. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a telecom support operation. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

B:

The suggested action—skip governance if output is fluent—does not establish the required control for a telecom support operation to document model limitations and acceptable use before regulated deployment. It replaces a verifiable engineering control with an untested shortcut in the a telecom support operation deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—ask end users to discover limitations only after incidents—does not establish the required control for a telecom support operation to document model limitations and acceptable use before regulated deployment. The change is applied outside the component responsible for the requirement identified by a telecom support operation. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 284

At a public-sector document office, engineers need to ingest recorded calls into a text-searchable knowledge collection. The next release must be evaluated against the previous baseline. The initial assessment isolates this issue from unrelated infrastructure changes. Which architecture is best supported by the stated constraints?

  1. Ask a text-only model to read binary recordings directly
  2. Use speech transcription, validate transcripts and chunk approved text for indexing
  3. Store only audio files in a text embedding index
  4. Assume OCR converts spoken audio to text

Explanation

Correct Answer

B:

Audio-to-text preprocessing makes recorded speech available to text-centric retrieval. In a public-sector document office, the engineering team should verify this design against its stated objective to ingest recorded calls into a text-searchable knowledge collection, including the requirement that the next release must be evaluated against the previous baseline.

Incorrect Answers

A:

The suggested action—ask a text-only model to read binary recordings directly—does not establish the required control for a public-sector document office to ingest recorded calls into a text-searchable knowledge collection. It replaces a verifiable engineering control with an untested shortcut in the a public-sector document office deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—store only audio files in a text embedding index—does not establish the required control for a public-sector document office to ingest recorded calls into a text-searchable knowledge collection. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a public-sector document office. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—assume OCR converts spoken audio to text—does not establish the required control for a public-sector document office to ingest recorded calls into a text-searchable knowledge collection. It replaces a verifiable engineering control with an untested shortcut in the a public-sector document office deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 285

At a clinical scheduling platform, engineers need to require machine-readable responses for downstream schema validation. The implementation must remain maintainable across model upgrades. A recent review found that the current design has no reliable safeguard for this case. What should be implemented before scaling this workflow?

  1. Specify an output schema and validate the parsed response before downstream actions
  2. Assume all natural-language answers are already valid JSON
  3. Skip validation because the prompt requests JSON
  4. Increase response randomness to guarantee deterministic fields

Explanation

Correct Answer

A:

Structured generation still needs explicit schema enforcement at integration boundaries. In a clinical scheduling platform, the engineering team should verify this design against its stated objective to require machine-readable responses for downstream schema validation, including the requirement that the implementation must remain maintainable across model upgrades.

Incorrect Answers

B:

The suggested action—assume all natural-language answers are already valid JSON—does not establish the required control for a clinical scheduling platform to require machine-readable responses for downstream schema validation. It replaces a verifiable engineering control with an untested shortcut in the a clinical scheduling platform deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—skip validation because the prompt requests JSON—does not establish the required control for a clinical scheduling platform to require machine-readable responses for downstream schema validation. The change is applied outside the component responsible for the requirement identified by a clinical scheduling platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—increase response randomness to guarantee deterministic fields—does not establish the required control for a clinical scheduling platform to require machine-readable responses for downstream schema validation. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a clinical scheduling platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 286

At a pharmaceutical documentation team, engineers need to recover from a temporary model throttle during API invocation. The team must preserve existing identity and authorization boundaries. The team wants a solution that can be validated through targeted test cases. Which architecture is best supported by the stated constraints?

  1. Treat throttling as proof that the model's factual answer is wrong
  2. Apply capped exponential backoff with jitter and respect quotas
  3. Retry in a tight infinite loop
  4. Increase all request payloads to improve quota allocation

Explanation

Correct Answer

B:

Rate-limit recovery needs paced retries and a strategy for sustained overload. In a pharmaceutical documentation team, the engineering team should verify this design against its stated objective to recover from a temporary model throttle during API invocation, including the requirement that the team must preserve existing identity and authorization boundaries.

Incorrect Answers

A:

The suggested action—treat throttling as proof that the model's factual answer is wrong—does not establish the required control for a pharmaceutical documentation team to recover from a temporary model throttle during API invocation. The change is applied outside the component responsible for the requirement identified by a pharmaceutical documentation team. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—retry in a tight infinite loop—does not establish the required control for a pharmaceutical documentation team to recover from a temporary model throttle during API invocation. It replaces a verifiable engineering control with an untested shortcut in the a pharmaceutical documentation team deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—increase all request payloads to improve quota allocation—does not establish the required control for a pharmaceutical documentation team to recover from a temporary model throttle during API invocation. The change is applied outside the component responsible for the requirement identified by a pharmaceutical documentation team. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 287

At a digital publishing business, engineers need to maintain service during a model-specific regional disruption. The implementation must remain maintainable across model upgrades. The initial assessment isolates this issue from unrelated infrastructure changes. What should be implemented before scaling this workflow?

  1. Disable data residency checks when invoking fallback Regions
  2. Move all customer records to a public web endpoint
  3. Evaluate eligible cross-Region inference profiles and compliance requirements
  4. Assume any model can be invoked in every destination Region

Explanation

Correct Answer

C:

Cross-Region choices depend on supported model profiles and allowed processing Regions. In a digital publishing business, the engineering team should verify this design against its stated objective to maintain service during a model-specific regional disruption, including the requirement that the implementation must remain maintainable across model upgrades.

Incorrect Answers

A:

The suggested action—disable data residency checks when invoking fallback Regions—does not establish the required control for a digital publishing business to maintain service during a model-specific regional disruption. It replaces a verifiable engineering control with an untested shortcut in the a digital publishing business deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

B:

The suggested action—move all customer records to a public web endpoint—does not establish the required control for a digital publishing business to maintain service during a model-specific regional disruption. The change is applied outside the component responsible for the requirement identified by a digital publishing business. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—assume any model can be invoked in every destination Region—does not establish the required control for a digital publishing business to maintain service during a model-specific regional disruption. It replaces a verifiable engineering control with an untested shortcut in the a digital publishing business deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 288

At a public-sector document office, engineers need to audit which users accessed sensitive retrieval documents. Response reliability is more important than free-form creativity. The team wants a solution that can be validated through targeted test cases. Which architecture is best supported by the stated constraints?

  1. Publish document access logs to anonymous users
  2. Enforce data-source authorization and emit protected access logs
  3. Use a single shared principal with no user attribution
  4. Rely solely on answer citations as an audit trail

Explanation

Correct Answer

B:

Auditable data access requires attributable identities and controlled log retention. In a public-sector document office, the engineering team should verify this design against its stated objective to audit which users accessed sensitive retrieval documents, including the requirement that response reliability is more important than free-form creativity.

Incorrect Answers

A:

The suggested action—publish document access logs to anonymous users—does not establish the required control for a public-sector document office to audit which users accessed sensitive retrieval documents. The change is applied outside the component responsible for the requirement identified by a public-sector document office. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—use a single shared principal with no user attribution—does not establish the required control for a public-sector document office to audit which users accessed sensitive retrieval documents. It replaces a verifiable engineering control with an untested shortcut in the a public-sector document office deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—rely solely on answer citations as an audit trail—does not establish the required control for a public-sector document office to audit which users accessed sensitive retrieval documents. The change is applied outside the component responsible for the requirement identified by a public-sector document office. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 289

At a SaaS procurement platform, engineers need to detect and redact personal identifiers before sending user documents to inference. The solution must tolerate transient dependency failures. The team wants a solution that can be validated through targeted test cases. Which remediation addresses the root problem?

  1. Send raw identifiers and ask the model to forget them afterward
  2. Use an S3 bucket name as a data redaction control
  3. Log entire unmasked prompts to every debugging dashboard
  4. Classify PII and apply approved redaction or masking before invocation

Explanation

Correct Answer

D:

Data minimization reduces exposure before sensitive information enters the model interaction. In a SaaS procurement platform, the engineering team should verify this design against its stated objective to detect and redact personal identifiers before sending user documents to inference, including the requirement that the solution must tolerate transient dependency failures.

Incorrect Answers

A:

The suggested action—send raw identifiers and ask the model to forget them afterward—does not establish the required control for a SaaS procurement platform to detect and redact personal identifiers before sending user documents to inference. It replaces a verifiable engineering control with an untested shortcut in the a SaaS procurement platform deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

B:

The suggested action—use an S3 bucket name as a data redaction control—does not establish the required control for a SaaS procurement platform to detect and redact personal identifiers before sending user documents to inference. The change is applied outside the component responsible for the requirement identified by a SaaS procurement platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

C:

The suggested action—log entire unmasked prompts to every debugging dashboard—does not establish the required control for a SaaS procurement platform to detect and redact personal identifiers before sending user documents to inference. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a SaaS procurement platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

Question 290

At a legal case-management provider, engineers need to support document lineage for generated answers. The architecture must minimize avoidable data exposure. An architecture review has rejected fixes that only conceal the symptoms. Which implementation most directly satisfies the requirement?

  1. Use anonymous embeddings with no reference to original documents
  2. Ask the language model to invent document revision IDs
  3. Store stable source identifiers, revision timestamps and permissions alongside chunks
  4. Discard all source identifiers after vectorization

Explanation

Correct Answer

C:

Chunk metadata supports traceability, filtering and source updates. In a legal case-management provider, the engineering team should verify this design against its stated objective to support document lineage for generated answers, including the requirement that the architecture must minimize avoidable data exposure.

Incorrect Answers

A:

The suggested action—use anonymous embeddings with no reference to original documents—does not establish the required control for a legal case-management provider to support document lineage for generated answers. It replaces a verifiable engineering control with an untested shortcut in the a legal case-management provider deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

B:

The suggested action—ask the language model to invent document revision IDs—does not establish the required control for a legal case-management provider to support document lineage for generated answers. The change is applied outside the component responsible for the requirement identified by a legal case-management provider. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

D:

The suggested action—discard all source identifiers after vectorization—does not establish the required control for a legal case-management provider to support document lineage for generated answers. It replaces a verifiable engineering control with an untested shortcut in the a legal case-management provider deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.

AWS Certified Generative AI Developer - Professional AIP-C01 Practice Tests