Q01 - Question
An accounts payable team receives a high volume of supplier invoices. They want a Power Automate flow to extract invoice data automatically so that they can process it without manual entry. They want to prototype quickly without training a model on their own data. Which Power Platform AI feature should you propose?
Domain: Design AI-powered business solutions Type: Single choice
- A. An AI Builder prebuilt model for receipt and invoice extraction
- B. Copilot Studio to build a custom conversational agent
- C. Copilot in Power Pages to generate forms from natural language
- D. Copilot in Power Apps to generate app screens from natural language
A is correct.
Explanation: AI Builder prebuilt models are useful for rapid prototyping or high-volume automation, and specifically include receipt and invoice extraction.
B is incorrect: Copilot Studio is used to build custom conversational agents with multi-turn reasoning, not for automated document extraction in a flow.
C is incorrect: Copilot in Power Pages is used to build site pages, forms, and data models with natural language, not for extracting data from invoices.
D is incorrect: Copilot in Power Apps is used to generate app screens, tables, and logic from natural language, not for document extraction.
Q02 - Question
A company is building a RAG solution on Azure AI Search to ground an agent in product documentation. The architect must maximize retrieval relevance and control index size and cost. Which two design choices should the architect make? Each correct answer presents part of the solution.
Domain: Design AI-powered business solutions Type: Multiple choice
- A. Query the live product systems on every conversation turn instead of building an index.
- B. Mark fields as searchable, filterable, sortable, or retrievable only when queries use them.
- C. Enable every capability on every field so that the schema supports any future query.
- D. Use hybrid retrieval that combines vector queries, keyword search, and semantic reranking.
B and D are correct.
Explanation: Extra field capabilities increase index size and cost, so mark fields only with the capabilities you use. Hybrid search, which combines vector, keyword, and semantic reranking, generally yields the most reliable relevance across phrasing, synonyms, and exact terms.
A is incorrect: Do not query live systems for every turn. Build an index that is optimized for your questions and updated on an SLO-driven freshness schedule.
C is incorrect: Enabling all capabilities increases storage and cost. Prefer a minimum viable schema.
Q03 - Question
A product owner asks you to approve a Microsoft 365 agent pilot. The agent will retrieve project files and write status updates to a tracking system through a connector. Before you approve the pilot, which two items should you confirm? Each correct answer presents part of the solution.
Domain: Design AI-powered business solutions Type: Multi-select
- A. The agent uses the largest available model so that output quality is as high as possible.
- B. The runs-as model is documented, and least-privilege access is confirmed.
- C. A custom agent is built first, and out-of-the-box agents are assessed after the custom build is complete.
- D. The required tools and connectors are identified, and failure paths and human approval points are defined.
B and D are correct.
Explanation: The readiness checklist for solution architects requires confirming identity and access (runs-as model documented, least-privilege confirmed) and actions and tools (required tools/connectors identified, failure paths and human approval points defined).
A is incorrect: The readiness checklist does not specify using the largest available model; it focuses on business value, identity, data scope, actions, security, change control, measurement, and support.
C is incorrect: The evidence states that before considering any custom agents, out-of-the-box pilots should be assessed to determine whether they will adequately meet the requirements.
Q04 - Question
A solution architect plans to extend Copilot in Dynamics 365 Sales with data from an external order management API. A developer proposes to build the custom connector in the tenant's default environment, which does not have Dynamics 365 apps enabled. What should the architect require?
Domain: Design AI-powered business solutions Type: Single choice
- A. Build the connector in the default environment and use an API key for authentication.
- B. Build the connector in the default environment, and then certify the connector and the plugin.
- C. Build the connector in an environment that has Dynamics 365 apps enabled, and use OAuth 2.0 with Microsoft Entra ID.
- D. Build the connector in the default environment, and then publish the Copilot action directly to Sales users without admin enablement.
C is correct.
Explanation: You must create the connector in an environment with Dynamics 365 apps enabled. Environments without Dynamics 365, such as the default environment, are not supported. Use OAuth 2.0 with Microsoft Entra ID as the identity provider for authentication.
A is incorrect: The default environment is not supported, and the authentication model for these connectors is OAuth 2.0 with Microsoft Entra ID, not an API key.
B is incorrect: Certification makes a connector available organization-wide. It does not remove the requirement to build the connector in an environment with Dynamics 365 apps enabled.
D is incorrect: The default environment is not supported, and an admin must enable the published action before it is visible to Sales users.
Q05 - Question
A finance organization designs an MCP-enabled agent in Copilot Studio for Dynamics 365 Finance & Operations. Auditors must be able to review the decisions the agent makes. Each user must only reach the business data they are allowed to access. Which design should you recommend?
Domain: Design AI-powered business solutions Type: Single choice
- A. Connect the agent to MCP context by using a shared account that can access all business entities.
- B. Rely on agent instructions to enforce responsible AI behavior, and do not log agent decisions.
- C. Expose all available data entities and domain models through MCP so the agent has complete context.
- D. Govern data access by user identity with least privilege, align MCP context boundaries with compliance controls, and log agent decisions.
D is correct.
Explanation: For MCP-enabled agents, govern data access by user identity by using least privilege. Align MCP context boundaries with compliance controls, and log agent decisions for auditability. This design meets both the access and audit requirements.
A is incorrect: A shared account with broad access does not limit data by user identity. It does not follow least privilege.
B is incorrect: Instructions that enforce responsible AI behavior are needed, but they do not replace logging. Auditors cannot review decisions that are not logged.
C is incorrect: Exposing all context ignores the requirement to align MCP context boundaries with compliance controls.
Q06 - Question
You are designing a Copilot Studio agent for an operations team. When a new row is added to a Dataverse table, the agent must send a summary to the team without waiting for a user to send a message. Which mechanism should you use?
Domain: Design AI-powered business solutions Type: Single choice
- A. A Generative Answers (NLU Boost) node
- B. A system topic
- C. A condition node
- D. An event trigger
D is correct.
Explanation: Event triggers let an agent act without waiting for a user prompt. A Dataverse row added is an example trigger event, and the trigger can start actions such as sending a summary. Triggers must be explicitly authorized and may impact billing consumption.
A is incorrect: Use generative answers when a user question is not covered by topics and the agent needs to answer from knowledge sources. It still responds to a user question.
B is incorrect: System topics handle common conversation situations, such as greeting, fallback, escalation, and errors. They are not designed to start work from a Dataverse event.
C is incorrect: Condition nodes branch a dialog based on variables and values. They do not start the agent when an event occurs.
Q07 - Question
Procurement users want to ask the Dynamics 365 Finance and Operations Copilot agent three things in one chat: whether a vendor is approved, when the vendor's latest compliance certificate expires, and to update the vendor's payment terms. Compliance certificates are stored in a SharePoint library. Which design should you recommend?
Domain: Design AI-powered business solutions Type: Single choice
- A. Ground the agent on Finance and Operations data only, because system data can answer all vendor questions.
- B. Use the SharePoint library as the source for all three requests, including the payment terms update.
- C. Create a client plugin that writes payment terms without checking Finance and Operations security roles, to reduce latency.
- D. Use Finance and Operations data for vendor approval, SharePoint for the certificate, and a Copilot client plugin that respects Finance and Operations security roles for the payment terms update.
D is correct.
Explanation: The evidence demonstrates mapping business questions to knowledge sources: F&O data for vendor status, SharePoint for compliance certificates, and plugin actions for updates. It also specifies that plugins must respect F&O security roles.
A is incorrect: F&O data alone may not answer all questions, as the compliance certificates are stored externally in a SharePoint library.
B is incorrect: SharePoint is an external knowledge base for retrieving documents, not a mechanism for executing actions like updating payment terms.
C is incorrect: The evidence explicitly states under governance and security that plugins must respect F&O security roles.
Q08 - Question
A financial services company is designing a Copilot Studio agent for a regulated account process. Users send consistent, structured commands such as "Reset password" and "Check balance." The business requires strict intent matching and deterministic responses with limited generative behavior. Which language understanding approach should you recommend?
Domain: Design AI-powered business solutions Type: Single choice
- A. Generative AI orchestration
- B. Standard natural language processing (NLU)
- C. Azure Conversational Language Understanding (CLU)
- D. A Generative Answers (NLU Boost) node as the primary routing method
B is correct.
Explanation: Use standard NLU for predictable tasks that require high precision and low variability. It fits strict intent matching for regulated processes, consistent structured commands, and deterministic responses with limited generative behavior.
A is incorrect: Generative orchestration is best for unstructured or unpredictable messages and open-ended tasks. It is also more compute intensive, which adds cost without benefit for fixed commands.
C is incorrect: Azure CLU is the best choice when phrasing varies moderately within defined topic boundaries. For structured commands with regulatory accuracy needs, standard NLU is the best choice.
D is incorrect: Generative answers act as a fallback when no topic matches. They answer from knowledge sources and do not provide strict, deterministic intent routing.
Q09 - Question
A support team uses Agent Builder in Microsoft 365 Copilot to create a case triage agent. The team has defined the agent's purpose, added instructions and guardrails, connected data sources, and configured actions and permissions. Leadership wants the agent available to support staff soon. What should the team do next?
Domain: Design AI-powered business solutions Type: Single choice
- A. Test the agent with sample prompts, and then publish it to targeted users or groups.
- B. Publish the agent to the entire organization, and then use user feedback to find errors.
- C. Add capabilities for policy Q&A and lead preparation so the agent serves more teams before release.
- D. Remove the escalation conditions so the agent can complete more cases without human review.
A is correct.
Explanation: In the Agent Builder workflow, you test the agent with sample prompts after you configure actions and permissions. You then publish the agent to targeted users or groups. Use the testing and validation workspace to verify behavior before release.
B is incorrect: Publishing to everyone skips the test step. Agent Builder includes publishing controls so you can target specific users or groups.
C is incorrect: Agents should solve a single high-value need. Adding unrelated capabilities makes intent less clear.
D is incorrect: Agent behavior should define conditions that escalate to human review. Removing them weakens the guardrails for case triage.
Q10 - Question
A utility company designs a Copilot Studio agent in voice mode to support field technicians. Technicians call the agent hands-free to request parts orders and check job status. Some requests trigger actions in downstream systems. Which behavior design should you recommend?
Domain: Design AI-powered business solutions Type: Single choice
- A. Provide detailed, multi-sentence spoken responses so technicians receive full context on each request.
- B. Use short, clear phrasing, provide confirmation steps, and include confidence checks before the agent executes actions.
- C. Execute actions immediately after the agent interprets intent so technicians do not wait for confirmation.
- D. Enable deep reasoning for every voice request so all responses go through structured multi-step evaluation.
B is correct.
Explanation: Voice agents should use short, clear phrasing, provide confirmation steps, and include confidence checks before executing actions. Voice relies on real-time interaction, so design for responsiveness, fallback handling, and user-friendly error messages.
A is incorrect: Avoid long multi-sentence responses in voice mode. The agent should provide concise spoken output.
C is incorrect: The voice interaction flow includes a step where the agent confirms or clarifies intent before it executes the action. Skipping this step increases the risk of incorrect actions.
D is incorrect: Deep reasoning is in preview and targets complex multi-step tasks. Standard reasoning provides fast responses and lower compute usage, which fits routine, high-volume requests such as job status checks.