Required Configuration Elements for Microsoft Foundry SDK Applications

AI-901 By islever Editorial Last updated

Implement generative AI apps and agents by using Foundry

Short answer

ADE You must provide the model deployment name, Microsoft Entra ID credentials, and the endpoint URL to authenticate and connect to the correct model instance via the Foundry SDK.

Question

You are developing a simple application that uses the Microsoft Foundry SDK to send chat prompts to a model.
Which three elements should you provide in the code to run the application? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
  1. the model deployment name Correct Answer
  2. the task type
  3. the Foundry project display name
  4. Microsoft Entra ID credentials Correct Answer
  5. the endpoint URL Correct Answer

Key Takeaways

The question tests the distinction between operational API requirements (credentials, URL, deployment name) and metadata or inference parameters (display name, task type).

Learn the three essential elements needed to run a Microsoft Foundry SDK application: model deployment name, Entra ID credentials, and endpoint URL.

Common mistake: Candidates often incorrectly select 'task type' or 'project display name', confusing human-readable labels or inference settings with the fundamental connection parameters required by the SDK.

Expert Analysis

To run a generative AI application using the Microsoft Foundry SDK, you must establish three critical components: identity, connectivity, and target. First, Microsoft Entra ID credentials (Option D) are mandatory for authentication; without valid credentials, such as DefaultAzureCredential, the SDK cannot authorize requests to the Azure AI services endpoint. Second, you need the endpoint URL (Option E), which specifies the exact API address (typically following the pattern <resource-name>.services.ai.azure.com) where the request is sent. Third, you must specify the model deployment name (Option A), which identifies the specific pre-deployed model instance within that project that will process the chat prompts.

The scenario describes sending chat prompts to a deployed model, not configuring the model itself. Therefore, the task type (Option B) is incorrect because task types (such as text-generation or image-generation) are generally inferred from the model architecture or specified in the prompt payload, not as a top-level configuration parameter for the connection. Similarly, the Foundry project display name (Option C) is a metadata label used for human readability in the Azure portal, not a technical identifier required for programmatic API access. The SDK relies on the unique resource identifiers (URL and deployment name) rather than human-readable names to route traffic correctly.

Exam Strategy

When configuring cloud AI SDKs, always look for the triad of authentication, connectivity, and target resource. Credentials verify who you are, the endpoint tells the system where to send data, and the deployment name tells the system which specific model to use. Ignore options related to human-readable labels (like display names) or high-level semantic tags (like task types) unless the question specifically asks about prompt engineering or model selection logic.

Frequently Asked Questions

Why is the task type not required in the code?

Task types are usually determined by the model's architecture or inferred from the prompt content. The SDK connects to a specific deployment, which inherently supports a specific modality like text generation.

Can I use the project name instead of the endpoint URL?

No. The project display name is for identification in the UI. The SDK requires the technical endpoint URL to route HTTP requests to the correct Azure resource.

More AI-901 FAQ →

Community Votes

A
34%
D
33%
E
33%

34% of anonymous learners picked answer A. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 1 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 1

  1. GrandWizard_007 👍 1
    A. Model deployment name - Identifies which deployed model your application will send chat prompts to. D. Microsoft Entra ID credentials - Used for authentication via DefaultAzureCredential or another Entra-based credential provider. Required for the SDK to authorize API calls. E. The endpoint URL - The Foundry project endpoint in the format: https://<resource-name>.services.ai.azure.com/api/projects/<project-name> Required for the SDK client to connect to your Foundry project.

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

OpenAI Images API Image Generation Code

S-grade

Azure Language PII Detection for Data Masking

S-grade

Azure Speech SDK Connection Configuration

S-grade

Microsoft Responsible AI Inclusiveness Principle

S-grade

Practice All AI-901 Questions

Access 105 questions with complete answers and detailed explanations.

View Full AI-901 Practice Test →

← Back to AI-901 Study Guide