McAfee Secure

Exam Code: DP-800

Exam Name: Developing AI-Enabled Database Solutions

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: SQL AI Developer Associate

Microsoft DP-800 Questions & Answers

Study with Up-To-Date REAL Exam Questions and Answers from the ACTUAL Test

101 Questions & Answers with Testing Engine

"Developing AI-Enabled Database Solutions Exam", also known as DP-800 exam, is a Microsoft certification exam.

Pass your tests with the always up-to-date DP-800 Exam Engine. Your DP-800 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Microsoft DP-800 Practice Test 77

Question 760

An engineer applies RRF to scores normalized per request and calls it a rank-based algorithm. Which technical action BEST resolves the primary problem?

  1. Use RRF to reward documents ranking highly across multiple candidate lists.
  2. Use exact KNN with a distance expression and top-k ordering.
  3. Evaluate indexed ANN search with measured recall/latency tradeoffs.
  4. Compute ranks within each result list before applying reciprocal-rank contributions.

Explanation

Correct Answer

D:

Compute ranks within each result list before applying reciprocal-rank contributions. This targets the primary issue because RRF operates on list positions and is distinct from raw score interpolation. Verification: A simple manual calculation matches fusion output. Prevent recurrence: Keep rank offset and truncation policies documented.

Incorrect Answers

A:

Using RRF to reward documents ranking highly across multiple candidate lists addresses a different database behavior; the relevant distinction is that compute ranks within each result list before applying reciprocal-rank contributions.

B:

Exact search compares the candidate vectors directly without ANN recall tradeoffs.

C:

Approximate indexes trade exact completeness for faster retrieval on large corpora.

Question 761

An API times out due to blocked UPDATE requests, but a team proposes adding more CPU. Which test result BEST demonstrates that the intended solution is working?

  1. The blocking chain is eliminated and update throughput recovers.
  2. A concurrent workload no longer produces avoidable deadlock victims.
  3. A forced or corrected plan improves tested workload without wrong results.
  4. A stale client receives a conflict and cannot overwrite the newer row.

Explanation

Correct Answer

A:

The blocking chain is eliminated and update throughput recovers. This is the decisive observable evidence after implementing identify the blocking head using supported DMV and wait analysis before scaling. It tests the underlying requirement: Lock waits are not resolved merely by allocating faster compute.

Incorrect Answers

B:

A concurrent workload no longer produces avoidable deadlock victims. However, that result would not establish the required outcome: the blocking chain is eliminated and update throughput recovers.

C:

A forced or corrected plan improves tested workload without wrong results. However, that result would not establish the required outcome: the blocking chain is eliminated and update throughput recovers.

D:

A stale client receives a conflict and cannot overwrite the newer row. However, that result would not establish the required outcome: the blocking chain is eliminated and update throughput recovers.

Question 762

An application needs a reusable deterministic tax rounding expression within a SQL query. Which technical principle BEST explains the relevant design decision?

  1. T-SQL UDFs have restrictions on side effects and cannot be arbitrary transactional commands.
  2. A scalar function produces one value for each call but may affect query optimization.
  3. Parameterization separates user values from executable SQL syntax.
  4. A view centralizes a query contract without storing a second copy by default.

Explanation

Correct Answer

B:

A scalar function produces one value for each call but may affect query optimization. This explains why the corrective action is to implement a typed scalar UDF only where its benefits outweigh inline expression overhead. Confirm the diagnosis when boundary amounts return correct DECIMAL values in unit tests.

Incorrect Answers

A:

T-SQL UDFs have restrictions on side effects and cannot be arbitrary transactional commands. The actual technical constraint is that a scalar function produces one value for each call but may affect query optimization.

C:

Parameterization separates user values from executable SQL syntax. The actual technical constraint is that a scalar function produces one value for each call but may affect query optimization.

D:

A view centralizes a query contract without storing a second copy by default. The actual technical constraint is that a scalar function produces one value for each call but may affect query optimization.

Question 763

An external model response contains invented order totals that differ from the authoritative SQL aggregation. Which ongoing control is MOST relevant after the issue is fixed?

  1. Implement safe fallback or manual review.
  2. Monitor truncation and fall back to SQL aggregates for precise counts.
  3. Preserve source IDs and show uncertainty where values cannot be verified.
  4. Use managed identity/secure credentials and enforce timeouts.

Explanation

Correct Answer

C:

Preserve source IDs and show uncertainty where values cannot be verified. This guards against recurrence because language model fluency does not establish numerical correctness. The direct remediation is: Reconcile generated claims with SQL source values before publishing results.

Incorrect Answers

A:

Implement safe fallback or manual review. The current task instead requires this action: preserve source IDs and show uncertainty where values cannot be verified.

B:

Monitor truncation and fall back to SQL aggregates for precise counts.

D:

Use managed identity/secure credentials and enforce timeouts. The current task instead requires this action: preserve source IDs and show uncertainty where values cannot be verified.

Question 764

An engineer assumes DACPAC build success proves production matches the project. What is the MOST appropriate immediate corrective approach?

  1. Compare live database schema against the project model before approval.
  2. Use dedicated scoped workload identity with least-privilege resource grants.
  3. Make required test/security checks and approval gates nonbypassable by normal contributors.
  4. Run schema comparison or drift reporting against the target environment.

Explanation

Correct Answer

D:

Run schema comparison or drift reporting against the target environment. This targets the primary issue because a valid model build cannot verify current deployed state. Verification: A test environment with modified index triggers a drift report. Prevent recurrence: Define drift policy for emergency fixes.

Incorrect Answers

A:

Manual changes create drift between intended declarative source and deployed objects.

B:

Using dedicated scoped workload identity with least-privilege resource grants would not correct an engineer assumes DACPAC build success proves production matches the project. Run schema comparison or drift reporting against the target environment.

C:

Making required test/security checks and approval gates nonbypassable by normal contributors would not correct an engineer assumes DACPAC build success proves production matches the project. Run schema comparison or drift reporting against the target environment.

Question 765

A manager needs a hierarchy of employees under one department head. What observed evidence would BEST confirm successful remediation?

  1. A small hierarchy returns all reachable descendants once under expected rules.
  2. A cyclic test input is detected rather than consuming unbounded resources.
  3. The test input produces distinct reviewer rows and correct identifiers.
  4. Test values return the intended code or an explicit missing result.

Explanation

Correct Answer

A:

A small hierarchy returns all reachable descendants once under expected rules. This is the decisive observable evidence after implementing use a recursive CTE with an anchor member and a terminating recursive member. It tests the underlying requirement: Hierarchical traversal requires recursion across parent-child relationships.

Incorrect Answers

B:

A cyclic test input is detected rather than consuming unbounded resources. However, that result would not establish the required outcome: a small hierarchy returns all reachable descendants once under expected rules.

C:

The test input produces distinct reviewer rows and correct identifiers. However, that result would not establish the required outcome: a small hierarchy returns all reachable descendants once under expected rules.

D:

Test values return the intended code or an explicit missing result. However, that result would not establish the required outcome: a small hierarchy returns all reachable descendants once under expected rules.

Question 766

A support report should obscure account numbers for junior analysts while allowing supervisors to see them. What is the MOST relevant technical distinction in this situation?

  1. Authorized query transformations and inference can expose sensitive values despite masking.
  2. DDM changes what unauthorized queries display; it is not data encryption.
  3. Database roles and schema grants can bound an automation's change surface.
  4. Read filtering alone does not necessarily stop unauthorized DML that changes row ownership.

Explanation

Correct Answer

B:

DDM changes what unauthorized queries display; it is not data encryption. This explains why the corrective action is to apply dynamic data masking with appropriate UNMASK grants and test role behavior. Confirm the diagnosis when low-privilege results show masked account numbers while privileged tests show permitted values.

Incorrect Answers

A:

Authorized query transformations and inference can expose sensitive values despite masking. The actual technical constraint is that DDM changes what unauthorized queries display; it is not data encryption.

C:

Database roles and schema grants can bound an automation's change surface. The actual technical constraint is that DDM changes what unauthorized queries display; it is not data encryption.

D:

Read filtering alone does not necessarily stop unauthorized DML that changes row ownership. The actual technical constraint is that DDM changes what unauthorized queries display; it is not data encryption.

Question 767

A production DAB release introduces a breaking API path but clients continue calling the old route. Which ongoing control is MOST relevant after the issue is fixed?

  1. Use telemetry and abuse alerting across API and SQL.
  2. Version the API contract and publish schema changes responsibly.
  3. Keep deployment artifacts and change history.
  4. Use environment-based connection references instead of inline secrets.

Explanation

Correct Answer

C:

Keep deployment artifacts and change history. This guards against recurrence because API endpoint changes can break consumers even if the underlying database remains healthy. The direct remediation is: Use versioned configuration and a staged deployment/rollback strategy.

Incorrect Answers

A:

Use telemetry and abuse alerting across API and SQL. The current task instead requires this action: keep deployment artifacts and change history.

B:

Version the API contract and publish schema changes responsibly. The current task instead requires this action: keep deployment artifacts and change history.

D:

Use environment-based connection references instead of inline secrets. The current task instead requires this action: keep deployment artifacts and change history.

Question 768

A helpdesk query seeks an exact phrase but FREETEXT returns loosely related documents. What is the MOST appropriate immediate corrective approach?

  1. Create a supported full-text index and required unique key on the searchable table.
  2. Use supported VECTORPROPERTY to obtain relevant vector metadata.
  3. Use reciprocal rank fusion with a consistent rank offset constant.
  4. Use a suitable CONTAINS phrase predicate for literal phrase intent.

Explanation

Correct Answer

D:

Use a suitable CONTAINS phrase predicate for literal phrase intent. This targets the primary issue because FREETEXT expands linguistic meaning and is not equivalent to exact phrase matching. Verification: The quoted phrase test excludes documents without that phrase. Prevent recurrence: Choose exact versus linguistic search deliberately.

Incorrect Answers

A:

Full-text predicates require a configured index and supported column types.

B:

Vector metadata functions describe representation without parsing serialized JSON arrays.

C:

RRF combines ordinal ranks rather than assuming score scales match.

Question 769

An order system accepts duplicate order numbers and downstream reconciliation double-counts orders. What observed evidence would BEST confirm successful remediation?

  1. A duplicate test insert fails while valid distinct orders succeed.
  2. An insert omitting the field records an acceptable UTC creation time.
  3. Rollback tests demonstrate permitted gaps without violating key uniqueness.
  4. Invalid payloads are rejected into an error queue without corrupting accepted rows.

Explanation

Correct Answer

A:

A duplicate test insert fails while valid distinct orders succeed. This is the decisive observable evidence after implementing enforce an appropriate UNIQUE constraint on the business order key after cleaning duplicates. It tests the underlying requirement: A surrogate primary key alone does not protect a separate business identifier.

Incorrect Answers

B:

An insert omitting the field records an acceptable UTC creation time. However, that result would not establish the required outcome: a duplicate test insert fails while valid distinct orders succeed.

C:

Rollback tests demonstrate permitted gaps without violating key uniqueness. However, that result would not establish the required outcome: a duplicate test insert fails while valid distinct orders succeed.

D:

Invalid payloads are rejected into an error queue without corrupting accepted rows. However, that result would not establish the required outcome: a duplicate test insert fails while valid distinct orders succeed.

Question 770

A pipeline deploys to production immediately after any branch push. Which technical reason BEST explains the proposed correction?

  1. Path ownership helps route high-impact DDL for accountable review.
  2. Automated execution without authorization can promote unreviewed schema changes.
  3. Repeatable deployments must not assume reference rows are absent.
  4. A text merge can compile but still change business behavior.

Explanation

Correct Answer

B:

Automated execution without authorization can promote unreviewed schema changes. This explains why the corrective action is to use protected branch triggers, environment approval gates and deployment-specific identities. Confirm the diagnosis when unapproved branch commits never reach production.

Incorrect Answers

A:

Path ownership helps route high-impact DDL for accountable review. The actual technical constraint is that automated execution without authorization can promote unreviewed schema changes.

C:

Repeatable deployments must not assume reference rows are absent. The actual technical constraint is that automated execution without authorization can promote unreviewed schema changes.

D:

A text merge can compile but still change business behavior. The actual technical constraint is that automated execution without authorization can promote unreviewed schema changes.

Developing AI-Enabled Database Solutions (DP-800) Practice Tests