Prepstellar
Get it on Google Play

AI-103 · Practice set 1 of 13

What Is Microsoft Foundry?: 10 practice questions

10 questions · Untimed · Free

10 free AI-103 practice questions on What Is Microsoft Foundry?, with an explanation for every answer. Untimed. The full mock exam and the timed version are in the app.

Set 1 · What Is Microsoft Foundry? · 10 questions Read the lesson

  1. Question 1 of 10

    In a Foundry resource, which level is the development boundary where a team builds and evaluates its use cases?

    1. AA project inside the Foundry resource
    2. BA connected Azure AI Search resource
    3. CA model deployment shared by teams
    4. DThe Foundry resource itself
    Show the answer

    Think of two layers. The Foundry resource is where IT governs networking, security, and model deployments. Inside it, each project is the workspace where a team builds and evaluates, reusing deployments and connections that already exist.

    Next → 1 / 10

  2. Question 2 of 10

    The services that were previously branded Azure AI Services now appear under which name?

    1. AFoundry Toolbox
    2. BFoundry Models
    3. CFoundry Tools
    4. DFoundry Agent Service
    Show the answer

    Separate the platform brand from the prebuilt-service brand. The first names the portal and resource; the second is the family containing Speech, Vision, and Language.

    Next → 2 / 10

  3. Question 3 of 10

    A product manager wants to explore models, prototype prompts, and build a prompt agent without writing code. Which developer surface fits best?

    1. AThe Azure Developer CLI (azd)
    2. BVisual Studio Code with the Foundry extension
    3. CThe Python SDK
    4. DThe Foundry portal
    Show the answer

    Match the surface to the person. No code at all points to the portal, which is built for exploring models, prototyping prompts, and building prompt agents. The CLI, VS Code, and the SDKs all assume someone is writing or running code.

    Next → 3 / 10

  4. Question 4 of 10

    A team works inside a Foundry project and wants to call the Translator API. At which scope is that API available?

    1. AOnly through a connected Azure AI Search resource
    2. BOnly at the project scope, like most new APIs
    3. CAt both the project and the resource scope
    4. DOnly at the Foundry resource level
    Show the answer

    Most new APIs live at the project scope, so it is easy to assume they all do. Some capabilities that came from the older services stay at the resource level, and Translator is the standard example: it is reached only from the Foundry resource.

    Next → 4 / 10

  5. Question 5 of 10

    Which of these actions is a control plane action in Foundry?

    1. ABuilding an agent
    2. BUploading files to a project
    3. CRunning an evaluation on a model
    4. DCreating a new project
    Show the answer

    Split actions by what they change. Control plane actions shape the setup itself, such as creating deployments and projects. Data plane actions are the daily work inside that setup: building agents, running evaluations, and uploading files.

    Next → 5 / 10

  6. Keep the ones you got wrong

    In the app, every question you miss comes back exactly when you’re about to forget it.

    Get it on Google Play

  7. Question 6 of 10

    How is a Foundry project represented in Azure Resource Manager?

    1. AAs a hub that groups Azure OpenAI and AI Services
    2. BAs a separate Microsoft.CognitiveServices/accounts resource
    3. CAs Microsoft.CognitiveServices/accounts/projects, a subresource
    4. DAs a Microsoft.Search resource joined by a connection
    Show the answer

    The Foundry resource is Microsoft.CognitiveServices/accounts with the kind AIServices. A project is not a second account: it is the accounts/projects subresource that lives under that Foundry resource.

    Next → 6 / 10

  8. Question 7 of 10

    A company upgrades its Azure OpenAI resource to a Foundry resource. What happens to its existing custom Azure policies and Azure RBAC actions?

    1. AThey continue to apply after the upgrade
    2. BThey must be recreated under a new provider namespace
    3. CThey move to the Foundry (classic) hub of the resource
    4. DThey are replaced by Foundry User
    Show the answer

    Foundry resources use the same Microsoft.CognitiveServices provider namespace as Azure OpenAI, so management APIs, RBAC actions, and policy aliases line up. That is why existing policies and RBAC actions keep working, and the endpoint, keys, and state are preserved too.

    Next → 7 / 10

  9. Question 8 of 10

    Three product teams need separate workspaces for their agents and evaluations. They must share the same model deployments, and central IT must control networking and security in one place. Which design fits?

    1. AOne project in one Foundry resource for all teams
    2. BOne Foundry resource with a project for each team
    3. CA separate Foundry resource for each team
    4. DA standalone Azure OpenAI resource for each team
    Show the answer

    Read the three constraints together. Separate workspaces calls for projects, shared deployments means one place that owns them, and central control means one governance layer. A single Foundry resource holding a project per team meets all three at once.

    Next → 8 / 10

  10. Question 9 of 10

    A compliance team must lock down network access to the Storage account, Key Vault, and Azure AI Search that a Foundry resource uses through connections. Where are those settings managed?

    1. AIn the Foundry Toolbox that exposes the knowledge
    2. BOn each connected resource, separately
    3. CIn each project that uses those connections
    4. DOn the Foundry resource, which governs its connections
    Show the answer

    A connection is a reference, not ownership. Storage, Key Vault, and Azure AI Search are independent Azure resources with their own governance boundaries, so their networking and access policies are configured on each of them, apart from the Foundry resource.

    Next → 9 / 10

  11. Question 10 of 10

    A single developer wants to try models, build a first agent, and run evaluations, with as little setup as possible. Which starting point is recommended?

    1. AA hub-based project in the Foundry (classic) portal
    2. BOne Foundry resource with one project
    3. CA standalone Azure OpenAI resource
    4. DTwo Foundry resources, one for models and one for agents
    Show the answer

    One Foundry resource already bundles model access, agent hosting, and evaluation tooling, and one project is enough for one person. That is the recommended default for single-developer exploration.

    Next → 10 / 10

  12. You’ve finished this set

    That’s 10 questions on What Is Microsoft Foundry?. In the app the ones you miss come back exactly when you’re about to forget them.

    Back to the course Next lesson: Foundry Tools at a Glance Get it on Google Play
Get it on Google Play

The whole course, on your phone

Lessons you can read, audio you can listen to on the way to work, and practice that remembers what you got wrong.

Get it on Google Play

“This app is the ‘One Lake’ for preparing the certification exams. Really worth it.”

Abdelhak Y. Google Play review