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 70

Question 690

A GraphQL endpoint permits nested queries that return customer records from unauthorized tenants. What maintenance or governance control BEST reduces recurrence of this problem?

  1. Combine DAB permissions with SQL RLS where applicable.
  2. Rate-limit expensive query shapes and monitor access logs.
  3. Prevent arbitrary tenant context spoofing and test privileged access.
  4. Reset pooled connections and validate context on each request.

Explanation

Correct Answer

B:

Rate-limit expensive query shapes and monitor access logs. This guards against recurrence because graphQL relationship traversal can increase data exposure if access is too broad. The direct remediation is: Enforce tenant-aware authorization, bounded query complexity and back-end SQL row security.

Incorrect Answers

A:

Combine DAB permissions with SQL RLS where applicable. The current task instead requires this action: rate-limit expensive query shapes and monitor access logs.

C:

Prevent arbitrary tenant context spoofing and test privileged access. The current task instead requires this action: rate-limit expensive query shapes and monitor access logs.

D:

Reset pooled connections and validate context on each request. The current task instead requires this action: rate-limit expensive query shapes and monitor access logs.

Question 691

A GraphQL query must return a customer's orders and related items without manual N+1 calls. Which implementation decision BEST addresses this requirement?

  1. Expose the stored procedure via a restricted DAB stored-procedure entity with execute rights.
  2. Create and validate a supported DAB JSON configuration with connection and entity definitions.
  3. Configure DAB entity relationships with matching source/target keys.
  4. Expose an authorized DAB REST endpoint with approved methods and resource paths.

Explanation

Correct Answer

C:

Configure DAB entity relationships with matching source/target keys. This targets the primary issue because correct GraphQL relationship metadata enables nested retrieval of related records. Verification: The customer fixture returns exactly its own order items. Prevent recurrence: Enforce SQL row-level security against cross-customer traversal.

Incorrect Answers

A:

DAB can surface declared stored procedures without granting arbitrary query execution.

B:

DAB generates endpoints from explicit configuration rather than automatically exposing every table.

D:

Expose an authorized DAB REST endpoint with approved methods and resource paths. The relevant technical condition instead is that configure DAB entity relationships with matching source/target keys.

Question 692

A model-serving endpoint is reachable publicly with no strong caller authentication. Which measurable outcome BEST verifies that the requirement is satisfied?

  1. Model calls and retrieval calls each succeed only with required scopes.
  2. Two test principals receive appropriately different display values.
  3. A role-based test demonstrates restricted account values are masked.
  4. Anonymous calls fail while authorized SQL workloads succeed.

Explanation

Correct Answer

D:

Anonymous calls fail while authorized SQL workloads succeed. This is the decisive observable evidence after implementing require approved identity authentication and network controls for the inference endpoint. It tests the underlying requirement: An open endpoint may expose inputs, outputs or costly compute to unauthorized callers.

Incorrect Answers

A:

Model calls and retrieval calls each succeed only with required scopes. However, that result would not establish the required outcome: anonymous calls fail while authorized SQL workloads succeed.

B:

Two test principals receive appropriately different display values. However, that result would not establish the required outcome: anonymous calls fail while authorized SQL workloads succeed.

C:

A role-based test demonstrates restricted account values are masked. However, that result would not establish the required outcome: anonymous calls fail while authorized SQL workloads succeed.

Question 693

A developer publishes a production DACPAC using staging variables and connects to the wrong database. Which technical property or constraint MOST directly matters here?

  1. Mis-targeted deployments can damage production regardless of SQL correctness.
  2. Changing project source alone does not apply DDL to the deployed database.
  3. Schema reconciliation can propose destructive changes that threaten existing records.
  4. Cross-system contracts can fail even when individual SQL procedures are correct.

Explanation

Correct Answer

A:

Mis-targeted deployments can damage production regardless of SQL correctness. This explains why the corrective action is to bind deployment environment, endpoint and credentials to explicit approved pipeline controls. Confirm the diagnosis when preflight reports match the authorized database identity and schema.

Incorrect Answers

B:

Changing project source alone does not apply DDL to the deployed database. The actual technical constraint is that mis-targeted deployments can damage production regardless of SQL correctness.

C:

Schema reconciliation can propose destructive changes that threaten existing records. The actual technical constraint is that mis-targeted deployments can damage production regardless of SQL correctness.

D:

Cross-system contracts can fail even when individual SQL procedures are correct. The actual technical constraint is that mis-targeted deployments can damage production regardless of SQL correctness.

Question 694

New shipments lack a creation timestamp whenever one integration omits the field. Which preventive practice BEST addresses the underlying risk over time?

  1. Review cascading deletes for unintended financial data loss.
  2. Separate creation time from event time and validate clock policy.
  3. Check memory capacity and unsupported features before migration.
  4. Monitor source availability and credential rotation.

Explanation

Correct Answer

B:

Separate creation time from event time and validate clock policy. This guards against recurrence because a default supplies a value on omission, not a guarantee against explicit bad values. The direct remediation is: Configure a suitable DEFAULT constraint with a UTC expression when required by the schema.

Incorrect Answers

A:

Review cascading deletes for unintended financial data loss. The current task instead requires this action: separate creation time from event time and validate clock policy.

C:

Check memory capacity and unsupported features before migration. The current task instead requires this action: separate creation time from event time and validate clock policy.

D:

Monitor source availability and credential rotation. The current task instead requires this action: separate creation time from event time and validate clock policy.

Question 695

A model sometimes produces markdown instead of the requested JSON object for an automated workflow. What should the SQL developer do FIRST to address the specific issue?

  1. Reconcile generated claims with SQL source values before publishing results.
  2. Project only required columns into a controlled JSON payload.
  3. Validate structured model output against a strict schema and reject invalid responses.
  4. Serialize selected relational fields with FOR JSON or appropriate JSON_OBJECT support.

Explanation

Correct Answer

C:

Validate structured model output against a strict schema and reject invalid responses. This targets the primary issue because generated output is not guaranteed to follow instructions unless independently checked. Verification: Malformed output is quarantined and valid output produces typed records. Prevent recurrence: Implement safe fallback or manual review.

Incorrect Answers

A:

Reconciling generated claims with SQL source values before publishing results would not correct a model sometimes produces markdown instead of the requested JSON object for an automated workflow. Validate structured model output against a strict schema and reject invalid responses.

B:

Data minimization reduces prompt exposure and cost without losing relevant context.

D:

Structured JSON preserves field boundaries and escaping for downstream model prompts.

Question 696

A teammate ships malformed DAB JSON with a misspelled source object name. Which acceptance check is MOST directly related to this scenario?

  1. Staging resolves its own Azure SQL endpoint and local development still works.
  2. Page requests return stable bounded records without gaps.
  3. Partners read approved columns and cannot modify base tables.
  4. A staging start resolves each declared source without runtime errors.

Explanation

Correct Answer

D:

A staging start resolves each declared source without runtime errors. This is the decisive observable evidence after implementing run dab validate and a connection smoke test before publishing. It tests the underlying requirement: Syntactic JSON validity alone does not prove a mapped SQL object exists.

Incorrect Answers

A:

Staging resolves its own Azure SQL endpoint and local development still works. However, that result would not establish the required outcome: a staging start resolves each declared source without runtime errors.

B:

Page requests return stable bounded records without gaps. However, that result would not establish the required outcome: a staging start resolves each declared source without runtime errors.

C:

Partners read approved columns and cannot modify base tables. However, that result would not establish the required outcome: a staging start resolves each declared source without runtime errors.

Question 697

A team adds two retrieval methods but simply concatenates results and duplicates many documents. Which technical property or constraint MOST directly matters here?

  1. Independent retrieval branches can produce overlapping hits and uncalibrated scores.
  2. Lexical retrieval protects exact tokens while vectors address paraphrases.
  3. Embedding similarity can confuse near-looking product codes that must match exactly.
  4. Full-text predicates require a configured index and supported column types.

Explanation

Correct Answer

A:

Independent retrieval branches can produce overlapping hits and uncalibrated scores. This explains why the corrective action is to deduplicate candidates by stable source ID and apply a documented fusion strategy. Confirm the diagnosis when results contain one entry per source passage in a stable ranking.

Incorrect Answers

B:

Lexical retrieval protects exact tokens while vectors address paraphrases. The actual technical constraint is that independent retrieval branches can produce overlapping hits and uncalibrated scores.

C:

Embedding similarity can confuse near-looking product codes that must match exactly. The actual technical constraint is that independent retrieval branches can produce overlapping hits and uncalibrated scores.

D:

Full-text predicates require a configured index and supported column types. The actual technical constraint is that independent retrieval branches can produce overlapping hits and uncalibrated scores.

Question 698

An employee should be prevented from inserting a row assigned to a different department. Which preventive practice BEST addresses the underlying risk over time?

  1. Reset pooled connections and validate context on each request.
  2. Test every write operation and auditing requirements.
  3. Keep a tool allowlist and capture authenticated invocations.
  4. Rehearse key recovery and avoid distributing keys in scripts.

Explanation

Correct Answer

B:

Test every write operation and auditing requirements. This guards against recurrence because read filtering alone does not necessarily stop unauthorized DML that changes row ownership. The direct remediation is: Add an RLS block predicate in addition to any required filter predicate.

Incorrect Answers

A:

Reset pooled connections and validate context on each request. The current task instead requires this action: test every write operation and auditing requirements.

C:

Keep a tool allowlist and capture authenticated invocations. The current task instead requires this action: test every write operation and auditing requirements.

D:

Rehearse key recovery and avoid distributing keys in scripts. The current task instead requires this action: test every write operation and auditing requirements.

Question 699

A DBA enables RLS but bulk export uses a high privilege or bypass identity. What should the SQL developer do FIRST to address the specific issue?

  1. Derive tenant identity from verified authentication and tightly controlled session setup.
  2. Replace excessive privileges with minimum SELECT and no UNMASK entitlement.
  3. Verify RLS behavior under every service principal and prohibit unnecessary bypass roles.
  4. Create or map the appropriate principal and grant minimum SQL permissions.

Explanation

Correct Answer

C:

Verify RLS behavior under every service principal and prohibit unnecessary bypass roles. This targets the primary issue because security enforcement must be tested with the exact executing principal. Verification: An export returns only rows permitted to its assigned role. Prevent recurrence: Monitor service-principal grants and run negative authorization tests.

Incorrect Answers

A:

This action addresses a different case: An RLS policy uses SESSION_CONTEXT set directly from untrusted request text. For that case, session context is only safe as an authorization input when callers cannot impersonate other tenants. For this case, it does not directly correct the observed problem; the distinguishing issue is security enforcement must be tested with the exact executing principal. The specifically relevant immediate technical action is: Verify RLS behavior under every service principal and prohibit unnecessary bypass roles.

B:

Replacing excessive privileges with minimum SELECT and no UNMASK entitlement would not correct a DBA enables RLS but bulk export uses a high privilege or bypass identity. Verify RLS behavior under every service principal and prohibit unnecessary bypass roles.

D:

This action addresses a different case: A container app's managed identity can log in but cannot read the required database view. For that case, azure identity authentication and database authorization are distinct checks. For this case, it does not directly correct the observed problem; the distinguishing issue is security enforcement must be tested with the exact executing principal. The specifically relevant immediate technical action is: Verify RLS behavior under every service principal and prohibit unnecessary bypass roles.

Question 700

A model endpoint returns a JSON envelope but the application saves the entire envelope as the business answer. Which test result BEST demonstrates that the intended solution is working?

  1. Malformed output is quarantined and valid output produces typed records.
  2. The approved identity receives a 2xx response while an unauthorized one fails.
  3. A JSON parser reconstructs the intended order IDs and amounts.
  4. A fixture response yields the expected text without transport headers.

Explanation

Correct Answer

D:

A fixture response yields the expected text without transport headers. This is the decisive observable evidence after implementing parse the supported response schema and extract the correct generated content field. It tests the underlying requirement: HTTP response metadata and generated text are different parts of the payload.

Incorrect Answers

A:

Malformed output is quarantined and valid output produces typed records. However, that result would not establish the required outcome: a fixture response yields the expected text without transport headers.

B:

The approved identity receives a 2xx response while an unauthorized one fails. However, that result would not establish the required outcome: a fixture response yields the expected text without transport headers.

C:

A JSON parser reconstructs the intended order IDs and amounts. However, that result would not establish the required outcome: a fixture response yields the expected text without transport headers.

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