100% Latest & Updated Microsoft AB-620 Practice Test Questions, Exam Dumps & Verified Answers!
30 Days Free Updates, Instant Download!
Question 450
In an energy utility, azure telemetry collection threatens to log raw authentication headers from connector calls. After the team implements this action: Exclude tokens and headers while retaining operation IDs, status and latency. Which evidence BEST verifies the requirement?
- The failure-rate dashboard reflects throttled and timed-out calls, not only successes.
- The prompt executes on the approved model and returns a schema-compliant response.
- After optimization, the dependency duration falls and end-to-end agent response improves.
- Application Insights supports diagnosis without storing credentials.
Explanation
Correct Answer
D:Application Insights supports diagnosis without storing credentials. This result tests the effect of excluding tokens and headers while retaining operation IDs, status and latency.
Incorrect Answers
A:The monitoring dashboard tracks only successful requests, hiding tool timeouts and throttling. In that setting, “The failure-rate dashboard reflects throttled and timed-out calls, not only successes” is relevant; the required choice here is “Application Insights supports diagnosis without storing credentials”.
B:An agent needs a different model from the default for a custom structured summarization prompt. That condition could justify “The prompt executes on the approved model and returns a schema-compliant response”. Here, “Application Insights supports diagnosis without storing credentials” targets the needed outcome.
C:“After optimization, the dependency duration falls and end-to-end agent response improves” could be appropriate when application Insights shows long tool-call duration but almost no model-generation latency. The present task instead requires “Application Insights supports diagnosis without storing credentials”.
Question 451
In a public university, the search index uses a stale document snapshot after source pages are updated. Which solution design decision BEST addresses this requirement?
- Instrument failures and dependencies in Application Insights alongside successful operations.
- Instrument the solution with Application Insights telemetry for agent operations.
- Establish a controlled refresh or reindex process for the Azure AI Search knowledge source.
- Use supported Microsoft Foundry model catalog integration for the custom prompt.
Explanation
Correct Answer
C:Establish a controlled refresh or reindex process for the Azure AI Search knowledge source. Implement this design by schedule or trigger appropriate index updates and verify indexing completion.
Incorrect Answers
A:The monitoring dashboard tracks only successful requests, hiding tool timeouts and throttling. That condition could justify “Instrument failures and dependencies in Application Insights alongside successful operations”. Here, “Establish a controlled refresh or reindex process for the Azure AI Search knowledge source” targets the needed outcome.
B:“Instrument the solution with Application Insights telemetry for agent operations” could be appropriate when an agent tool call intermittently fails and operators need service-level traces. The present task instead requires “Establish a controlled refresh or reindex process for the Azure AI Search knowledge source”.
D:An agent needs a different model from the default for a custom structured summarization prompt. That condition could justify “Use supported Microsoft Foundry model catalog integration for the custom prompt”. Here, “Establish a controlled refresh or reindex process for the Azure AI Search knowledge source” targets the needed outcome.
Question 452
In a public university, the search index uses a stale document snapshot after source pages are updated. The team selected this design: Establish a controlled refresh or reindex process for the Azure AI Search knowledge source. Which implementation action BEST puts it into practice?
- Use correlated dependency telemetry to identify and optimize the slow API call.
- Configure approved telemetry, correlation IDs and safe log fields.
- Schedule or trigger appropriate index updates and verify indexing completion.
- Apply the user's authorized filter or permissions when retrieving indexed evidence.
Explanation
Correct Answer
C:Schedule or trigger appropriate index updates and verify indexing completion. The expected result is that a re-run retrieves the new document content rather than outdated text.
Incorrect Answers
A:“Use correlated dependency telemetry to identify and optimize the slow API call” could be appropriate when application Insights shows long tool-call duration but almost no model-generation latency. The present task instead requires “Schedule or trigger appropriate index updates and verify indexing completion”.
B:An agent tool call intermittently fails and operators need service-level traces. In that setting, “Configure approved telemetry, correlation IDs and safe log fields” is relevant; the required choice here is “Schedule or trigger appropriate index updates and verify indexing completion”.
D:“Apply the user's authorized filter or permissions when retrieving indexed evidence” could be appropriate when a search-backed agent returns data from Azure AI Search but needs access control to follow each requester's identity. The present task instead requires “Schedule or trigger appropriate index updates and verify indexing completion”.
Question 453
In a public university, the search index uses a stale document snapshot after source pages are updated. After the team implements this action: Schedule or trigger appropriate index updates and verify indexing completion. Which evidence BEST verifies the requirement?
- Reviewers can see whether evidence was missing or the model ignored available evidence.
- A re-run retrieves the new document content rather than outdated text.
- Each grounded answer can be traced to the retrieved article.
- Prompt traces contain the necessary appointment fields but not unrelated medical data.
Explanation
Correct Answer
B:A re-run retrieves the new document content rather than outdated text. This result tests the effect of schedule or trigger appropriate index updates and verify indexing completion.
Incorrect Answers
A:The agent must distinguish model hallucination from missing retrieval evidence when an answer is wrong. In that setting, “Reviewers can see whether evidence was missing or the model ignored available evidence” is relevant; the required choice here is “A re-run retrieves the new document content rather than outdated text”.
C:“Each grounded answer can be traced to the retrieved article” could be appropriate when an agent generates answers against Azure AI Search results but omits document attribution. The present task instead requires “A re-run retrieves the new document content rather than outdated text”.
D:A custom Foundry prompt accepts raw patient data even though it needs only appointment type. In that setting, “Prompt traces contain the necessary appointment fields but not unrelated medical data” is relevant; the required choice here is “A re-run retrieves the new document content rather than outdated text”.
Question 454
In a retail payments team, a custom Foundry model prompt is switched without checking safety and accuracy regression. Which solution design decision BEST addresses this requirement?
- Instrument failures and dependencies in Application Insights alongside successful operations.
- Use supported Microsoft Foundry model catalog integration for the custom prompt.
- Treat model changes as evaluated solution changes rather than invisible configuration tweaks.
- Apply version filtering and document metadata to Azure AI Search grounding.
Explanation
Correct Answer
C:Treat model changes as evaluated solution changes rather than invisible configuration tweaks. Implement this design by running representative test cases on the new model and reviewing result differences before promotion.
Incorrect Answers
A:The monitoring dashboard tracks only successful requests, hiding tool timeouts and throttling. That condition could justify “Instrument failures and dependencies in Application Insights alongside successful operations”. Here, “Treat model changes as evaluated solution changes rather than invisible configuration tweaks” targets the needed outcome.
B:“Use supported Microsoft Foundry model catalog integration for the custom prompt” could be appropriate when an agent needs a different model from the default for a custom structured summarization prompt. The present task instead requires “Treat model changes as evaluated solution changes rather than invisible configuration tweaks”.
D:Search indexes contain multiple versions of content and the agent cites an obsolete policy. That condition could justify “Apply version filtering and document metadata to Azure AI Search grounding”. Here, “Treat model changes as evaluated solution changes rather than invisible configuration tweaks” targets the needed outcome.
Question 455
In a retail payments team, a custom Foundry model prompt is switched without checking safety and accuracy regression. The team selected this design: Treat model changes as evaluated solution changes rather than invisible configuration tweaks. Which implementation action BEST puts it into practice?
- Configure the prompt/model resource and validate output shape and usage permissions.
- Use approved configuration values to point production prompts at a valid supported deployment.
- Run representative test cases on the new model and review result differences before promotion.
- Configure approved telemetry, correlation IDs and safe log fields.
Explanation
Correct Answer
C:Run representative test cases on the new model and review result differences before promotion. The expected result is that the candidate model meets approved correctness, safety and latency gates.
Incorrect Answers
A:“Configure the prompt/model resource and validate output shape and usage permissions” could be appropriate when an agent needs a different model from the default for a custom structured summarization prompt. The present task instead requires “Run representative test cases on the new model and review result differences before promotion”.
B:A Foundry model deployment used in development does not exist in the production subscription. In that setting, “Use approved configuration values to point production prompts at a valid supported deployment” is relevant; the required choice here is “Run representative test cases on the new model and review result differences before promotion”.
D:“Configure approved telemetry, correlation IDs and safe log fields” could be appropriate when an agent tool call intermittently fails and operators need service-level traces. The present task instead requires “Run representative test cases on the new model and review result differences before promotion”.
Question 456
In a retail payments team, a custom Foundry model prompt is switched without checking safety and accuracy regression. After the team implements this action: Run representative test cases on the new model and review result differences before promotion. Which evidence BEST verifies the requirement?
- The candidate model meets approved correctness, safety and latency gates.
- The failure-rate dashboard reflects throttled and timed-out calls, not only successes.
- After optimization, the dependency duration falls and end-to-end agent response improves.
- The prompt executes on the approved model and returns a schema-compliant response.
Explanation
Correct Answer
A:The candidate model meets approved correctness, safety and latency gates. This result tests the effect of running representative test cases on the new model and reviewing result differences before promotion.
Incorrect Answers
B:The monitoring dashboard tracks only successful requests, hiding tool timeouts and throttling. That condition could justify “The failure-rate dashboard reflects throttled and timed-out calls, not only successes”. Here, “The candidate model meets approved correctness, safety and latency gates” targets the needed outcome.
C:“After optimization, the dependency duration falls and end-to-end agent response improves” could be appropriate when application Insights shows long tool-call duration but almost no model-generation latency. The present task instead requires “The candidate model meets approved correctness, safety and latency gates”.
D:An agent needs a different model from the default for a custom structured summarization prompt. In that setting, “The prompt executes on the approved model and returns a schema-compliant response” is relevant; the required choice here is “The candidate model meets approved correctness, safety and latency gates”.
Question 457
In a banking support operation, an agent passes happy-path tests but fails on policy exceptions that occur in real conversations. Which solution design decision BEST addresses this requirement?
- Build a representative evaluation test set that includes routine and edge-case requests.
- Expand the evaluation set to supported languages and relevant local phrasing.
- Evaluate schema conformity as a distinct acceptance criterion.
- Measure repeatability and validate numerical correctness across repeated runs.
Explanation
Correct Answer
A:Build a representative evaluation test set that includes routine and edge-case requests. Implement this design by adding labeled cases for normal questions, policy exceptions and adversarial input with expected behavior.
Incorrect Answers
B:“Expand the evaluation set to supported languages and relevant local phrasing” could be appropriate when an agent fails in multilingual conversations but all current tests use English. The present task instead requires “Build a representative evaluation test set that includes routine and edge-case requests”.
C:A new prompt helps ordinary cases but breaks structured output parsing. In that setting, “Evaluate schema conformity as a distinct acceptance criterion” is relevant; the required choice here is “Build a representative evaluation test set that includes routine and edge-case requests”.
D:A new agent model responds inconsistently to the same deterministic calculation prompt. That condition could justify “Measure repeatability and validate numerical correctness across repeated runs”. Here, “Build a representative evaluation test set that includes routine and edge-case requests” targets the needed outcome.
Question 458
In a banking support operation, an agent passes happy-path tests but fails on policy exceptions that occur in real conversations. The team selected this design: Build a representative evaluation test set that includes routine and edge-case requests. Which implementation action BEST puts it into practice?
- Remove repeated scenarios and add independent booking intents and failure modes.
- Run both candidates on the same labeled inputs and compare per-case and aggregate outcomes.
- Score reference-supported answers against expected evidence as well as text quality.
- Add labeled cases for normal questions, policy exceptions and adversarial input with expected behavior.
Explanation
Correct Answer
D:Add labeled cases for normal questions, policy exceptions and adversarial input with expected behavior. The expected result is that test runs reveal exception failures instead of reporting only successful easy cases.
Incorrect Answers
A:“Remove repeated scenarios and add independent booking intents and failure modes” could be appropriate when an evaluation set contains duplicate questions and yields an overly optimistic success rate. The present task instead requires “Add labeled cases for normal questions, policy exceptions and adversarial input with expected behavior”.
B:The team wants to compare two agent versions on the same customer scenarios. In that setting, “Run both candidates on the same labeled inputs and compare per-case and aggregate outcomes” is relevant; the required choice here is “Add labeled cases for normal questions, policy exceptions and adversarial input with expected behavior”.
C:Agent answers must be correct and cite controlling regulatory documents. That condition could justify “Score reference-supported answers against expected evidence as well as text quality”. Here, “Add labeled cases for normal questions, policy exceptions and adversarial input with expected behavior” targets the needed outcome.
Question 459
In a banking support operation, an agent passes happy-path tests but fails on policy exceptions that occur in real conversations. After the team implements this action: Add labeled cases for normal questions, policy exceptions and adversarial input with expected behavior. Which evidence BEST verifies the requirement?
- The agent asks for corrections rather than inventing claim details.
- A correct-looking hallucinated status without a real tool call fails.
- Test runs reveal exception failures instead of reporting only successful easy cases.
- Auditors can reproduce which agent build and test definitions produced each score.
Explanation
Correct Answer
C:Test runs reveal exception failures instead of reporting only successful easy cases. This result tests the effect of adding labeled cases for normal questions, policy exceptions and adversarial input with expected behavior.
Incorrect Answers
A:An agent test set contains only valid claims, ignoring missing attachments and corrupt input files. In that setting, “The agent asks for corrections rather than inventing claim details” is relevant; the required choice here is “Test runs reveal exception failures instead of reporting only successful easy cases”.
B:A test suite checks only the final answer, not whether the correct backend was invoked. That condition could justify “A correct-looking hallucinated status without a real tool call fails”. Here, “Test runs reveal exception failures instead of reporting only successful easy cases” targets the needed outcome.
D:The team cannot prove what changed between agent evaluations across releases. In that setting, “Auditors can reproduce which agent build and test definitions produced each score” is relevant; the required choice here is “Test runs reveal exception failures instead of reporting only successful easy cases”.
Question 460
In a customer billing team, the team wants to compare two agent versions on the same customer scenarios. Which solution design decision BEST addresses this requirement?
- Evaluate out-of-scope safety and escalation behavior rather than only response fluency.
- Record version identifiers, test-set revision and scoring rubric with each run.
- Measure repeatability and validate numerical correctness across repeated runs.
- Use a fixed evaluation test set with consistent scoring criteria for both versions.
Explanation
Correct Answer
D:Use a fixed evaluation test set with consistent scoring criteria for both versions. Implement this design by running both candidates on the same labeled inputs and comparing per-case and aggregate outcomes.
Incorrect Answers
A:A test asks a symptom question and the agent responds as if issuing a clinical diagnosis. That condition could justify “Evaluate out-of-scope safety and escalation behavior rather than only response fluency”. Here, “Use a fixed evaluation test set with consistent scoring criteria for both versions” targets the needed outcome.
B:“Record version identifiers, test-set revision and scoring rubric with each run” could be appropriate when the team cannot prove what changed between agent evaluations across releases. The present task instead requires “Use a fixed evaluation test set with consistent scoring criteria for both versions”.
C:A new agent model responds inconsistently to the same deterministic calculation prompt. In that setting, “Measure repeatability and validate numerical correctness across repeated runs” is relevant; the required choice here is “Use a fixed evaluation test set with consistent scoring criteria for both versions”.
Designing and Building Integrated AI Agent Solutions in Copilot Studio (AB-620) Practice Tests