AI-103 Certification Practice Question #31

Question

You are building a web app named App1 that generates responses by using a model deployed to a Microsoft Foundry project named Project1.
Before sending the prompts to the model, App1 must retrieve documents by using Azure AI Search.
You need to integrate Project1 and App1. The solution must meet the following requirements:
Multiple client applications must use the same search configuration.
A security policy must prevent key-based authentication.
Administrative effort must be minimized.
What should you do?

Accepted Answer

A project-level Azure AI Search connection in Foundry is created once and referenced by every client application, so all apps share the same search configuration (requirement 1). The connection uses the project's Microsoft Entra managed identity for keyless authentication, satisfying the policy that bans key-based auth (requirement 2), and it is centrally managed in the project so administrative effort is minimized (requirement 3). Calling Search directly from each app (C, D) or building per-app custom HTTP connections (A) duplicates configuration and authentication, failing the "same configuration / minimal effort" requirements. MS Learn documents exactly this connection-plus-managed-identity pattern as the recommended way to ground Foundry agents/apps on an Azure AI Search index.

More AI-103 practice questions