azure-ai-projects

In the Ultra Transcenders books

AI-103

Each book explains azure-ai-projects in context, with comparison tables and the common traps.

Terms in this definition

• SDK

  Software development kit: client libraries, for example the Key Vault SDK or Speech SDK, that let code use a service without calling its REST API directly.

• Microsoft Foundry

  Formerly called Azure AI Foundry, the Azure platform where AI models and agents are built, evaluated and run within one resource, with evaluations, guardrails and AI red teaming.

• AIProjectClient

  Starting object of the Microsoft Foundry SDK, constructed from the project endpoint plus a Microsoft Entra credential like DefaultAzureCredential. Because only Entra ID authentication is supported, no project key or SAS exists.

• Connection

  Resource that attaches a virtual network gateway to its peer, which may be an ExpressRoute circuit, a second VNet gateway (Vnet2Vnet) or a local network gateway over IPsec. Resetting it recovers a single tunnel and avoids rebooting the whole gateway.

• Foundry project

  A container beneath a Foundry resource that keeps one team's agents, data, files, evaluations and project connections separate. Model deployments and resource-level connections are shared from the parent, and governance is not configured at this level.

• Application Insights

  APM service within Azure Monitor providing application telemetry, availability tests, usage analytics and Application Map. Data from workspace-based instances lives in Log Analytics.

Related terms

• azure-core-tracing-opentelemetry

  Python package connecting Azure SDK calls to OpenTelemetry; for tracing agents on the client side it is installed together with opentelemetry-sdk and azure-ai-projects.

• Foundry SDK

  Client library (azure-ai-projects) that operates on a Microsoft Foundry project via AIProjectClient, covering agents, datasets, deployments and connections. It also provides an OpenAI-compatible client for conversations, responses and evaluations.