Snowflake SnowPro Specialty Gen AI GES-C01 Practice Test 27
Question 260
At a regional airline services firm, a Snowflake engineering team is undertaking an incident summary assistant. Natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. The proposed fix is still hypothetical; audit leads demand source-grounded conclusions, and network outages create intermittent retries. Which TWO checks provide the most decisive evidence before approving a change?
- Compare key obligations and dates across languages
- Test ambiguous aliases explicitly.
- Compare generated columns to intended dimension
- Identify which input produced the failure
- Measure mean input tokens and completions per job
Explanation
Correct Answers
B:This complementary observation tests a different part of the same hypothesis and should be read together with the first check before authorizing change. Expected success would be: questions use the approved region dimension and validated filters. These author-intended interpretations require independent technical review before operational use.
C:This observation directly tests the case mechanism: Column-name guessing can select a similar but inequivalent attribute. In the described setting, it helps determine whether add governed semantic synonyms and representative verified query examples for regions. is justified. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
A:Checking this might help diagnose language translation risk, where a cross-border service desk automatically translates legal notices before human review. It is not one of the two most direct checks for the present case because the leading mechanism is column-name guessing can select a similar but inequivalent attribute. These author-intended interpretations require independent technical review before operational use.
D:Checking this might help diagnose try_complete failure isolation, where a scheduled text pipeline stops processing whenever one model call fails on a problematic input. It is not one of the two most direct checks for the present case because the leading mechanism is column-name guessing can select a similar but inequivalent attribute. These author-intended interpretations require independent technical review before operational use.
E:Checking this might help diagnose cortex complete budget, where an overnight summarization task sends full transcripts and consumes far more credits than the pilot. It is not one of the two most direct checks for the present case because the leading mechanism is column-name guessing can select a similar but inequivalent attribute. These author-intended interpretations require independent technical review before operational use.
Question 261
At a public transit agency, a Snowflake engineering team is undertaking a fleet operations question-answering tool. Natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Diagnosis is disputed because the data includes safety incident narratives; changes require accountable signoff. Which underlying failure mechanism BEST explains the observed limitation?
- Treating each LLM call as infallible makes one transient failure break an entire batch.
- A dominant sample can conceal contrary trends in smaller groups.
- Column-name guessing can select a similar but inequivalent attribute.
- A model invocation cannot safely infer mismatched feature names or types.
Explanation
Correct Answer
C:This is the causal hypothesis consistent with the observed limitation: natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Confirm it through both compare generated columns to intended dimension and test ambiguous aliases explicitly. before changing the service. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
A:That mechanism could explain a different case about try_complete failure isolation, but does not most directly account for this specific symptom: natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Here, the evidence to collect is compare generated columns to intended dimension These author-intended interpretations require independent technical review before operational use.
B:That mechanism could explain a different case about sentiment aggregation bias, but does not most directly account for this specific symptom: natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Here, the evidence to collect is compare generated columns to intended dimension These author-intended interpretations require independent technical review before operational use.
D:That mechanism could explain a different case about model registry signature mismatch, but does not most directly account for this specific symptom: natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Here, the evidence to collect is compare generated columns to intended dimension These author-intended interpretations require independent technical review before operational use.
Question 262
At a global food wholesaler, a Snowflake engineering team is undertaking a vendor invoice reconciliation. Natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. The change was trialed on a small dataset; accounting must prove each extracted total, while suppliers use incompatible document formats. Which acceptance condition BEST demonstrates that the intervention worked?
- Questions use the approved region dimension and validated filters.
- Failed rows are quarantined while valid rows finish and are auditable.
- Batches meet extraction quality at an agreed cost per item.
- Inference happens only through approved geographic paths.
Explanation
Correct Answer
A:This is the observable acceptance condition that would show the intended fix addressed column-name guessing can select a similar but inequivalent attribute. It is stronger than activity counts because it tests the user-facing effect. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
B:This could demonstrate success for try_complete failure isolation, but it would not establish resolution of the current issue: natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Validate questions use the approved region dimension and validated filters. instead. These author-intended interpretations require independent technical review before operational use.
C:This could demonstrate success for cortex complete budget, but it would not establish resolution of the current issue: natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Validate questions use the approved region dimension and validated filters. instead. These author-intended interpretations require independent technical review before operational use.
D:This could demonstrate success for cross-region inference decision, but it would not establish resolution of the current issue: natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Validate questions use the approved region dimension and validated filters. instead. These author-intended interpretations require independent technical review before operational use.
Question 263
At a university research office, a Snowflake engineering team is undertaking a restricted data discovery project. Natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. An auditor will inspect the control design; research access differs by grant, and staff rotate between roles during the academic year. Which ongoing control MOST directly reduces the original risk?
- Version synonyms with business-definition review.
- Pin deployable semantic models and preserve rollback copies.
- Add bounded retries and a dead-letter review path.
- Add fairness and sampling checks before business decisions.
Explanation
Correct Answer
A:This retains a specific safeguard against recurrence of column-name guessing can select a similar but inequivalent attribute. It complements the immediate intervention, add governed semantic synonyms and representative verified query examples for regions., and allows the team to verify the intended result. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
B:This is a sensible safeguard for semantic yaml lineage, but it does not target the concrete exposure in this scenario. The current failure is column-name guessing can select a similar but inequivalent attribute. These author-intended interpretations require independent technical review before operational use.
C:This is a sensible safeguard for try_complete failure isolation, but it does not target the concrete exposure in this scenario. The current failure is column-name guessing can select a similar but inequivalent attribute. These author-intended interpretations require independent technical review before operational use.
D:This is a sensible safeguard for sentiment aggregation bias, but it does not target the concrete exposure in this scenario. The current failure is column-name guessing can select a similar but inequivalent attribute. These author-intended interpretations require independent technical review before operational use.
Question 264
At a community-bank technology team, a Snowflake engineering team is undertaking an internal policy assistant pilot. Natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Budget and operational risk are both material; security rejects any unapproved cross-border data processing; the pilot must serve two departments with different permissions. Which decision MOST appropriately balances the requested outcome with production constraints?
- Add governed semantic synonyms and representative verified query examples for regions. Then version synonyms with business-definition review.
- Enforce warehouse permissions for generated SQL and validate service identity privileges. Postpone the critical evidence checks until after a broad release.
- Aggregate with cohort and weighting choices explained rather than reporting only the pooled mean. Postpone the critical evidence checks until after a broad release.
- Profile retrieval latency separately from generation and tune search service design. Postpone the critical evidence checks until after a broad release.
Explanation
Correct Answer
A:This combines a corrective action with a bounded production safeguard, addressing natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. Prove the result by checking that questions use the approved region dimension and validated filters. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
B:This proposal begins with an intervention for cortex analyst access check and postpones evidence that is essential for the current case. Before rollout, the team should add governed semantic synonyms and representative verified query examples for regions. These author-intended interpretations require independent technical review before operational use.
C:This proposal begins with an intervention for sentiment aggregation bias and postpones evidence that is essential for the current case. Before rollout, the team should add governed semantic synonyms and representative verified query examples for regions. These author-intended interpretations require independent technical review before operational use.
D:This proposal begins with an intervention for cortex search latency and postpones evidence that is essential for the current case. Before rollout, the team should add governed semantic synonyms and representative verified query examples for regions. These author-intended interpretations require independent technical review before operational use.
Question 265
At a professional services consultancy, a Snowflake engineering team is undertaking a role-scoped analytics portal. Quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Before implementation, consultants work on multiple confidential clients; project billing must be attributable. Which first architectural or operating step MOST directly addresses this case?
- Summarize the source text with an explicit schema for decisions, risks, and unresolved actions.
- Configure structured output expectations and validate each returned object against an application schema.
- Add governed semantic synonyms and representative verified query examples for regions.
- Add approved SQL examples to the verified query collection and evaluate paraphrased requests.
Explanation
Correct Answer
D:The decision addresses the central constraint: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. The most direct next step is to add approved sql examples to the verified query collection and evaluate paraphrased requests. Before changing production, review both: reconcile result totals to certified reports and inspect sql for semantic drift.. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
A:This can be appropriate when analysts need concise meeting highlights without losing decisions, blockers, or action owners. It addresses summarization target, whereas the present limitation is better explained by without trusted examples, text-to-sql may vary materially under paraphrase. The target outcome here is that common requests reproduce the expected metric computation. These author-intended interpretations require independent technical review before operational use.
B:This can be appropriate when a downstream api rejects model replies because some fields are numeric in one record and strings in another. It addresses structured outputs schema, whereas the present limitation is better explained by without trusted examples, text-to-sql may vary materially under paraphrase. The target outcome here is that common requests reproduce the expected metric computation. These author-intended interpretations require independent technical review before operational use.
C:This can be appropriate when natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. It addresses cortex analyst schema ambiguity, whereas the present limitation is better explained by without trusted examples, text-to-sql may vary materially under paraphrase. The target outcome here is that common requests reproduce the expected metric computation. These author-intended interpretations require independent technical review before operational use.
Question 266
At an industrial maintenance company, a Snowflake engineering team is undertaking an asset fault knowledge base. Quarterly board reporting requires the same recognized calculation across different natural-language phrasings. The proposed fix is still hypothetical; field logs contain abbreviations and exact part numbers, and repair crews require concise results. Which TWO checks provide the most decisive evidence before approving a change?
- Compare generated columns to intended dimension
- Test whether injected text changes the assistant's role
- Inspect labels for overlapping examples
- Reconcile result totals to certified reports
- Inspect SQL for semantic drift.
Explanation
Correct Answers
D:This observation directly tests the case mechanism: Without trusted examples, text-to-SQL may vary materially under paraphrase. In the described setting, it helps determine whether add approved sql examples to the verified query collection and evaluate paraphrased requests. is justified. These author-intended interpretations require independent technical review before operational use.
E:This complementary observation tests a different part of the same hypothesis and should be read together with the first check before authorizing change. Expected success would be: common requests reproduce the expected metric computation. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
A:Checking this might help diagnose cortex analyst schema ambiguity, where natural-language queries repeatedly choose the wrong region field because customers use informal abbreviations. It is not one of the two most direct checks for the present case because the leading mechanism is without trusted examples, text-to-sql may vary materially under paraphrase. These author-intended interpretations require independent technical review before operational use.
B:Checking this might help diagnose complete prompt injection, where a helpdesk assistant uses a customer email body as context, and the email orders it to reveal confidential data. It is not one of the two most direct checks for the present case because the leading mechanism is without trusted examples, text-to-sql may vary materially under paraphrase. These author-intended interpretations require independent technical review before operational use.
C:Checking this might help diagnose classify_text boundary, where a support routing classifier assigns multiple tags where the workflow requires exactly one primary queue. It is not one of the two most direct checks for the present case because the leading mechanism is without trusted examples, text-to-sql may vary materially under paraphrase. These author-intended interpretations require independent technical review before operational use.
Question 267
At a renewable-energy service provider, a Snowflake engineering team is undertaking an engineering knowledge assistant. Quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Diagnosis is disputed because technicians work in intermittent connectivity conditions; the business needs verifiable source references. Which underlying failure mechanism BEST explains the observed limitation?
- A model invocation cannot safely infer mismatched feature names or types.
- Descriptive prompts alone do not guarantee stable types or required fields.
- Treating each LLM call as infallible makes one transient failure break an entire batch.
- Without trusted examples, text-to-SQL may vary materially under paraphrase.
Explanation
Correct Answer
D:This is the causal hypothesis consistent with the observed limitation: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Confirm it through both reconcile result totals to certified reports and inspect sql for semantic drift. before changing the service. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
A:That mechanism could explain a different case about model registry signature mismatch, but does not most directly account for this specific symptom: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Here, the evidence to collect is reconcile result totals to certified reports These author-intended interpretations require independent technical review before operational use.
B:That mechanism could explain a different case about structured outputs schema, but does not most directly account for this specific symptom: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Here, the evidence to collect is reconcile result totals to certified reports These author-intended interpretations require independent technical review before operational use.
C:That mechanism could explain a different case about try_complete failure isolation, but does not most directly account for this specific symptom: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Here, the evidence to collect is reconcile result totals to certified reports These author-intended interpretations require independent technical review before operational use.
Question 268
At a non-profit social services network, a Snowflake engineering team is undertaking a service quality dashboard. Quarterly board reporting requires the same recognized calculation across different natural-language phrasings. The change was trialed on a small dataset; the organization serves populations with different access constraints, while some participant data is confidential. Which acceptance condition BEST demonstrates that the intervention worked?
- Common requests reproduce the expected metric computation.
- Valid requests produce supported predictions without implicit type coercion surprises.
- Key negative complaints retain meaning across selected languages.
- The assistant answers from permitted context without obeying the embedded command.
Explanation
Correct Answer
A:This is the observable acceptance condition that would show the intended fix addressed without trusted examples, text-to-sql may vary materially under paraphrase. It is stronger than activity counts because it tests the user-facing effect. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
B:This could demonstrate success for model registry signature mismatch, but it would not establish resolution of the current issue: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Validate common requests reproduce the expected metric computation. instead. These author-intended interpretations require independent technical review before operational use.
C:This could demonstrate success for translation quality control, but it would not establish resolution of the current issue: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Validate common requests reproduce the expected metric computation. instead. These author-intended interpretations require independent technical review before operational use.
D:This could demonstrate success for complete prompt injection, but it would not establish resolution of the current issue: quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Validate common requests reproduce the expected metric computation. instead. These author-intended interpretations require independent technical review before operational use.
Question 269
At a packaging manufacturer, a Snowflake engineering team is undertaking a product support search interface. Quarterly board reporting requires the same recognized calculation across different natural-language phrasings. An auditor will inspect the control design; operators need reliable revision dates, and knowledge articles update every workday. Which ongoing control MOST directly reduces the original risk?
- Enforce classification contract and review uncertain tickets.
- Version synonyms with business-definition review.
- Regression-test the verified query set after schema changes.
- Set query timeouts and controlled result-size limits.
Explanation
Correct Answer
C:This retains a specific safeguard against recurrence of without trusted examples, text-to-sql may vary materially under paraphrase. It complements the immediate intervention, add approved sql examples to the verified query collection and evaluate paraphrased requests., and allows the team to verify the intended result. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
A:This is a sensible safeguard for classify_text boundary, but it does not target the concrete exposure in this scenario. The current failure is without trusted examples, text-to-sql may vary materially under paraphrase. These author-intended interpretations require independent technical review before operational use.
B:This is a sensible safeguard for cortex analyst schema ambiguity, but it does not target the concrete exposure in this scenario. The current failure is without trusted examples, text-to-sql may vary materially under paraphrase. These author-intended interpretations require independent technical review before operational use.
D:This is a sensible safeguard for sql query execution monitoring, but it does not target the concrete exposure in this scenario. The current failure is without trusted examples, text-to-sql may vary materially under paraphrase. These author-intended interpretations require independent technical review before operational use.
Question 270
At a semiconductor equipment maker, a Snowflake engineering team is undertaking a component troubleshooting assistant. Quarterly board reporting requires the same recognized calculation across different natural-language phrasings. Budget and operational risk are both material; document titles are duplicated across product revisions; engineers need verified part-specific guidance. Which decision MOST appropriately balances the requested outcome with production constraints?
- Add approved SQL examples to the verified query collection and evaluate paraphrased requests.
- Reject invalid payloads before writing business records.
- Require traceable excerpts for material action items.
- Regression-test the verified query set after schema changes.
- Version synonyms with business-definition review.
Explanation
Correct Answers
A:This implements the response to the stated failure: Without trusted examples, text-to-SQL may vary materially under paraphrase. It must be validated using reconcile result totals to certified reports These author-intended interpretations require independent technical review before operational use.
D:This establishes the ongoing safety boundary after the immediate correction. It provides protection while the team checks that common requests reproduce the expected metric computation. These author-intended interpretations require independent technical review before operational use.
Incorrect Answers
B:This guardrail primarily addresses structured outputs schema and is not the companion intervention the case most urgently requires. The stated issue is quarterly board reporting requires the same recognized calculation across different natural-language phrasings. These author-intended interpretations require independent technical review before operational use.
C:This guardrail primarily addresses summarization target and is not the companion intervention the case most urgently requires. The stated issue is quarterly board reporting requires the same recognized calculation across different natural-language phrasings. These author-intended interpretations require independent technical review before operational use.
E:This guardrail primarily addresses cortex analyst schema ambiguity and is not the companion intervention the case most urgently requires. The stated issue is quarterly board reporting requires the same recognized calculation across different natural-language phrasings. These author-intended interpretations require independent technical review before operational use.