Question 140
At a wholesale distribution platform, engineers need to deploy a large self-hosted model with significant GPU memory needs. The service must continue to support multiple independent tenants. The team wants a solution that can be validated through targeted test cases. Which remediation addresses the root problem?
- Treat model deployment as a CPU-only static web hosting task
- Assume endpoint throughput is independent of generated token length
- Run a large GPU workload in an API Gateway mapping template
- 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 wholesale distribution platform, 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 service must continue to support multiple independent tenants.
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 wholesale distribution platform 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 wholesale distribution platform. 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 wholesale distribution platform 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 wholesale distribution 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—run a large GPU workload in an API Gateway mapping template—does not establish the required control for a wholesale distribution platform 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 wholesale distribution platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 141
At a travel services company, engineers need to detect and redact personal identifiers before sending user documents to inference. The design must distinguish inference behavior from upstream data quality. A recent review found that the current design has no reliable safeguard for this case. Which design decision addresses the underlying issue?
- Classify PII and apply approved redaction or masking before invocation
- Send raw identifiers and ask the model to forget them afterward
- Use an S3 bucket name as a data redaction control
- Log entire unmasked prompts to every debugging dashboard
Explanation
Correct Answer
A:Data minimization reduces exposure before sensitive information enters the model interaction. In a travel services company, 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 design must distinguish inference behavior from upstream data quality.
Incorrect Answers
B:The suggested action—send raw identifiers and ask the model to forget them afterward—does not establish the required control for a travel services company 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 travel services company. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
C:The suggested action—use an S3 bucket name as a data redaction control—does not establish the required control for a travel services company 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 travel services company deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
D:The suggested action—log entire unmasked prompts to every debugging dashboard—does not establish the required control for a travel services company 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 travel services company. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 142
At a commercial property platform, engineers need to document model limitations and acceptable use before regulated deployment. The compliance team requires a traceable decision record. A recent review found that the current design has no reliable safeguard for this case. Which implementation most directly satisfies the requirement?
- Skip governance if output is fluent
- Ask end users to discover limitations only after incidents
- Maintain model cards, risk assessments and accountable approvals
- Treat a model vendor marketing page as the complete risk assessment
Explanation
Correct Answer
C:Governance artifacts define limitations, ownership and acceptance criteria. In a commercial property platform, 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—skip governance if output is fluent—does not establish the required control for a commercial property platform to document model limitations and acceptable use before regulated deployment. It replaces a verifiable engineering control with an untested shortcut in the a commercial property 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—ask end users to discover limitations only after incidents—does not establish the required control for a commercial property platform to document model limitations and acceptable use before regulated deployment. The change is applied outside the component responsible for the requirement identified by a commercial property platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
D:The suggested action—treat a model vendor marketing page as the complete risk assessment—does not establish the required control for a commercial property platform to document model limitations and acceptable use before regulated deployment. It replaces a verifiable engineering control with an untested shortcut in the a commercial property platform deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 143
At a wholesale distribution platform, engineers need to detect harmful output regressions before a prompt release. The architecture must minimize avoidable data exposure. The team wants a solution that can be validated through targeted test cases. What is the most appropriate production approach?
- Approve a change because compilation succeeds
- Skip tests after modifying a prompt instead of code
- Use overall application uptime as a safety score
- Run a fixed adversarial and safety evaluation suite as a CI/CD gate
Explanation
Correct Answer
D:Versioned safety cases detect policy regressions before users encounter them. In a wholesale distribution platform, the engineering team should verify this design against its stated objective to detect harmful output regressions before a prompt release, including the requirement that the architecture must minimize avoidable data exposure.
Incorrect Answers
A:The suggested action—approve a change because compilation succeeds—does not establish the required control for a wholesale distribution platform to detect harmful output regressions before a prompt release. It replaces a verifiable engineering control with an untested shortcut in the a wholesale distribution 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—skip tests after modifying a prompt instead of code—does not establish the required control for a wholesale distribution platform to detect harmful output regressions before a prompt release. The change is applied outside the component responsible for the requirement identified by a wholesale distribution platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
C:The suggested action—use overall application uptime as a safety score—does not establish the required control for a wholesale distribution platform to detect harmful output regressions before a prompt release. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a wholesale distribution platform. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 144
At a global manufacturer, engineers need to normalize inconsistent aliases before generating embeddings. The architecture must minimize avoidable data exposure. An architecture review has rejected fixes that only conceal the symptoms. Which change should the team prioritize?
- Increase top-p to compensate for source inconsistency
- Remove all metadata because the input is inconsistent
- Apply deterministic text normalization and entity mapping in the preprocessing pipeline
- Store inconsistent aliases without preserving a canonical mapping
Explanation
Correct Answer
C:Canonical preprocessing reduces semantically duplicate or mismatched indexed content. In a global manufacturer, the engineering team should verify this design against its stated objective to normalize inconsistent aliases before generating embeddings, including the requirement that the architecture must minimize avoidable data exposure.
Incorrect Answers
A:The suggested action—increase top-p to compensate for source inconsistency—does not establish the required control for a global manufacturer to normalize inconsistent aliases before generating embeddings. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a global manufacturer. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
B:The suggested action—remove all metadata because the input is inconsistent—does not establish the required control for a global manufacturer to normalize inconsistent aliases before generating embeddings. It replaces a verifiable engineering control with an untested shortcut in the a global manufacturer deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
D:The suggested action—store inconsistent aliases without preserving a canonical mapping—does not establish the required control for a global manufacturer to normalize inconsistent aliases before generating embeddings. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a global manufacturer. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 145
At a telecom support operation, engineers need to reject malformed or incomplete source records before RAG ingestion. The implementation must remain maintainable across model upgrades. A recent review found that the current design has no reliable safeguard for this case. Which remediation addresses the root problem?
- Raise the generation temperature to repair corrupted source data
- Implement schema and quality checks with quarantining and measurable failure thresholds
- Embed every record even if required fields are absent
- Silently discard rows while reporting full ingestion success
Explanation
Correct Answer
B:Data validation catches ingestion defects before they contaminate retrieval results. In a telecom support operation, the engineering team should verify this design against its stated objective to reject malformed or incomplete source records before RAG ingestion, including the requirement that the implementation must remain maintainable across model upgrades.
Incorrect Answers
A:The suggested action—raise the generation temperature to repair corrupted source data—does not establish the required control for a telecom support operation to reject malformed or incomplete source records before RAG ingestion. 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.
C:The suggested action—embed every record even if required fields are absent—does not establish the required control for a telecom support operation to reject malformed or incomplete source records before RAG ingestion. 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.
D:The suggested action—silently discard rows while reporting full ingestion success—does not establish the required control for a telecom support operation to reject malformed or incomplete source records before RAG ingestion. 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.
Question 146
At a municipal services portal, engineers need to debug malformed tool arguments emitted by a model. Operators need measurable acceptance criteria before rollout. The team wants a solution that can be validated through targeted test cases. What should be implemented before scaling this workflow?
- Increase vector search shards without inspecting API errors
- Ignore schema failures until the model guesses the right shape
- Validate against the tool schema, inspect failed payloads and refine tool definitions
- Execute malformed payloads with administrator rights
Explanation
Correct Answer
C:Tool-call diagnosis examines model outputs against executable interface contracts. In a municipal services portal, the engineering team should verify this design against its stated objective to debug malformed tool arguments emitted by a model, including the requirement that operators need measurable acceptance criteria before rollout.
Incorrect Answers
A:The suggested action—increase vector search shards without inspecting API errors—does not establish the required control for a municipal services portal to debug malformed tool arguments emitted by a model. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a municipal services portal. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
B:The suggested action—ignore schema failures until the model guesses the right shape—does not establish the required control for a municipal services portal to debug malformed tool arguments emitted by a model. It replaces a verifiable engineering control with an untested shortcut in the a municipal services 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—execute malformed payloads with administrator rights—does not establish the required control for a municipal services portal to debug malformed tool arguments emitted by a model. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a municipal services portal. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 147
At a digital publishing business, engineers need to decouple customer event ingestion from slow downstream GenAI processing. A controlled rollout is required before broad user exposure. The initial assessment isolates this issue from unrelated infrastructure changes. Which design decision addresses the underlying issue?
- Store event payloads in a prompt template as the message queue
- Retry in the user browser after it closes
- Publish an event and process asynchronously with durable retry or queue semantics
- Block the transaction until every AI job finishes
Explanation
Correct Answer
C:Event-driven integration separates business transaction latency from model processing. In a digital publishing business, the engineering team should verify this design against its stated objective to decouple customer event ingestion from slow downstream GenAI processing, including the requirement that a controlled rollout is required before broad user exposure.
Incorrect Answers
A:The suggested action—store event payloads in a prompt template as the message queue—does not establish the required control for a digital publishing business to decouple customer event ingestion from slow downstream GenAI processing. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a digital publishing business. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
B:The suggested action—retry in the user browser after it closes—does not establish the required control for a digital publishing business to decouple customer event ingestion from slow downstream GenAI processing. 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.
D:The suggested action—block the transaction until every AI job finishes—does not establish the required control for a digital publishing business to decouple customer event ingestion from slow downstream GenAI processing. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a digital publishing business. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 148
At a public-sector document office, engineers need to route easy requests to inexpensive models while preserving quality on difficult requests. The architecture must minimize avoidable data exposure. The team wants a solution that can be validated through targeted test cases. Which architecture is best supported by the stated constraints?
- Select a model solely by its parameter count
- Use a policy-driven model router with measurable fallback conditions
- Send all requests to the largest model regardless of complexity
- Change production source code manually for every routing decision
Explanation
Correct Answer
B:Capability-aware routing balances quality, latency and invocation cost. In a public-sector document office, the engineering team should verify this design against its stated objective to route easy requests to inexpensive models while preserving quality on difficult requests, including the requirement that the architecture must minimize avoidable data exposure.
Incorrect Answers
A:The suggested action—select a model solely by its parameter count—does not establish the required control for a public-sector document office to route easy requests to inexpensive models while preserving quality on difficult requests. 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—send all requests to the largest model regardless of complexity—does not establish the required control for a public-sector document office to route easy requests to inexpensive models while preserving quality on difficult requests. 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—change production source code manually for every routing decision—does not establish the required control for a public-sector document office to route easy requests to inexpensive models while preserving quality on difficult requests. 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 149
At a security incident response team, engineers need to improve time to first visible answer token in a chat interface. The migration must preserve existing downstream API contracts. The initial assessment isolates this issue from unrelated infrastructure changes. Which change should the team prioritize?
- Stream output and measure first-token latency separately from total completion time
- Wait until the entire completion is generated before displaying it
- Add unnecessary serial model calls before the first token
- Increase maximum response length as the latency remedy
Explanation
Correct Answer
A:Streaming improves perceived responsiveness by showing partial output early. In a security incident response team, the engineering team should verify this design against its stated objective to improve time to first visible answer token in a chat interface, including the requirement that the migration must preserve existing downstream API contracts.
Incorrect Answers
B:The suggested action—wait until the entire completion is generated before displaying it—does not establish the required control for a security incident response team to improve time to first visible answer token in a chat interface. The change is applied outside the component responsible for the requirement identified by a security incident response team. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
C:The suggested action—add unnecessary serial model calls before the first token—does not establish the required control for a security incident response team to improve time to first visible answer token in a chat interface. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by a security incident response team. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
D:The suggested action—increase maximum response length as the latency remedy—does not establish the required control for a security incident response team to improve time to first visible answer token in a chat interface. It replaces a verifiable engineering control with an untested shortcut in the a security incident response team deployment. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.
Question 150
At an automotive parts supplier, engineers need to audit which users accessed sensitive retrieval documents. The current deployment processes customer-specific information. The team wants a solution that can be validated through targeted test cases. Which architecture is best supported by the stated constraints?
- Publish document access logs to anonymous users
- Enforce data-source authorization and emit protected access logs
- Use a single shared principal with no user attribution
- Rely solely on answer citations as an audit trail
Explanation
Correct Answer
B:Auditable data access requires attributable identities and controlled log retention. In an automotive parts supplier, the engineering team should verify this design against its stated objective to audit which users accessed sensitive retrieval documents, including the requirement that the current deployment processes customer-specific information.
Incorrect Answers
A:The suggested action—publish document access logs to anonymous users—does not establish the required control for an automotive parts supplier to audit which users accessed sensitive retrieval documents. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by an automotive parts supplier. 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 an automotive parts supplier to audit which users accessed sensitive retrieval documents. The change is applied outside the component responsible for the requirement identified by an automotive parts supplier. 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 an automotive parts supplier to audit which users accessed sensitive retrieval documents. No demonstrated capability in this scenario guarantees that this step resolves the concern raised by an automotive parts supplier. In this deployment, engineers should verify the missing mechanism rather than assume the proposed step satisfies the acceptance conditions.