You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
description: This course introduces core concepts related to artificial intelligence (AI), and the services in Microsoft Azure that can be used to create AI solutions, focusing on Microsoft Foundry.
6
+
description: This course introduces core concepts related to artificial intelligence (AI), and the services in Microsoft Azure that can be used to create AI solutions, focusing on Microsoft Foundry.
6
7
brand: azure
7
8
ms.date: 12/11/2025
8
9
author: sherzyang
9
10
ms.author: sheryang
10
11
ms.topic: learning-path
11
12
title: "Introduction to AI in Azure"
12
13
summary: |
13
-
This course introduces core concepts related to artificial intelligence (AI), and the services in Microsoft Azure that can be used to create AI solutions.
14
+
This course introduces core concepts related to artificial intelligence (AI), and the services in Microsoft Azure that can be used to create AI solutions.
14
15
prerequisites: Familiarity with Azure and the Azure portal.
Copy file name to clipboardExpand all lines: learn-pr/wwl-data-ai/ai-information-extraction/index.yml
+16-16Lines changed: 16 additions & 16 deletions
Original file line number
Diff line number
Diff line change
@@ -1,5 +1,6 @@
1
1
### YamlMime:Module
2
2
uid: learn.wwl.ai-information-extraction
3
+
hidden: true
3
4
metadata:
4
5
title: (AI-900) Get started with AI-powered information extraction in Microsoft Foundry
5
6
description: AI gives you the power to unlock insights from your data. In this module, you'll learn how to use Foundry Tools to extract information from content.
title: (AI-900) Get started with Microsoft Foundry
5
6
description: "This module introduces Microsoft Foundry and its capabilities for building AI solutions. Explore the key features and benefits of using Azure for AI development."
@@ -14,24 +15,24 @@ abstract: Get started with AI in Microsoft Foundry
14
15
prerequisites: Ability to navigate the Azure portal
title: (AI-900) Get started with computer vision in Microsoft Foundry
5
6
description: Azure Vision service enables software engineers to create intelligent solutions that extract information from images; a common task in many artificial intelligence (AI) scenarios. Learn how to use the Azure Vision service to analyze images.
@@ -17,27 +18,26 @@ abstract: Learn how to use the Azure Vision service to analyze images.
17
18
prerequisites: Ability to navigate the Azure portal
title: (AI-900) Get started with generative AI and agents in Microsoft Foundry
5
6
description: "Learn how you can build generative AI applications and agents in Microsoft Foundry. Understand how to use Microsoft Foundry's extensive model marketplace, playgrounds, and evolving features."
@@ -10,30 +11,29 @@ metadata:
10
11
ms.topic: module
11
12
ai-usage: ai-assisted
12
13
title: (AI-900) Get started with generative AI and agents in Microsoft Foundry
13
-
summary: Learn how you can build generative AI applications and agents in Microsoft Foundry. Understand how to use Microsoft Foundry's extensive model marketplace, playgrounds, and evolving features.
14
+
summary: Learn how you can build generative AI applications and agents in Microsoft Foundry. Understand how to use Microsoft Foundry's extensive model marketplace, playgrounds, and evolving features.
14
15
abstract: Get started with generative AI and agents in Microsoft Foundry
title: (AI-900) Get started with natural language processing in Microsoft Foundry
5
6
description: "Explore Azure Language's text analysis features, which include sentiment analysis, key phrase extraction, named entity recognition, and language detection."
@@ -10,7 +11,7 @@ metadata:
10
11
ms.topic: module
11
12
ms.collection: wwl-ai-copilot
12
13
ms.custom:
13
-
- N/A
14
+
- N/A
14
15
ms.service: azure-ai-language
15
16
ai-usage: ai-assisted
16
17
title: (AI-900) Get started with natural language processing in Microsoft Foundry
@@ -19,27 +20,27 @@ abstract: Learn how to use Azure Language for text analysis
19
20
prerequisites: Ability to navigate the Azure portal
-**System instructions** – sets behavior and role of the model.
23
23
24
-
Unlike the user prompt, which is the end-user request or question (example: Where should I travel?), a **System prompt** sets behavior, tone, tools, and guardrails for the assistant. An example of a system prompt is: "You are a helpful, step‑by‑step tutor. Cite sources. Decline medical advice."
24
+
Unlike the user prompt, which is the end-user request or question (example: Where should I travel?), a **System prompt** sets behavior, tone, tools, and guardrails for the assistant. An example of a system prompt is: "You are a helpful, step‑by‑step tutor. Cite sources. Decline medical advice."
25
25
26
-
The playground is a useful bridge between Foundry and code. After you test representative prompts, you can use the same system and user prompts and parameter values in your code. The playground provides code that can call your Foundry deployment via the OpenAI‑compatible *Responses* API. The code is essentially what is running when you use the chat interface to configure settings and send user prompts.
26
+
The playground is a useful bridge between Foundry and code. After you test representative prompts, you can use the same system and user prompts and parameter values in your code. The playground provides code that can call your Foundry deployment via the OpenAI‑compatible *Responses* API. The code is essentially what is running when you use the chat interface to configure settings and send user prompts.
27
27
28
28
29
29
30
-
You can take the code as a starting point for creating your own chat client.
30
+
You can take the code as a starting point for creating your own chat client.
31
31
32
32
## Create a lightweight chat client by using the Foundry SDK
33
33
@@ -38,14 +38,11 @@ A **lightweight client application** is a small, minimal app whose primary job i
38
38
- Has a **small code footprint** and minimal configuration (often just environment variables + a short script).
39
39
- Is easy to prototype, easy to run locally, and easy to extend later.
40
40
41
-
For Foundry, a lightweight chat client is often a **single Python file** that connects to a Foundry project endpoint and sends chat messages to a deployed model. The Foundry SDK exposes a **Project client** (Foundry‑native ops) and an **OpenAI‑compatible client** for calling models via the **Responses API**. Most apps use both.
41
+
For Foundry, a lightweight chat client is often a **single Python file** that connects to a Foundry project endpoint and sends chat messages to a deployed model. The Foundry SDK exposes a **Project client** (Foundry‑native ops) and an **OpenAI‑compatible client** for calling models via the **Responses API**. Most apps use both.
42
42
43
43
#### Build a Python chat client
44
44
45
-
After you created a **Foundry project** and **deployed a chat model** (for example, `gpt-4.1`), you can use the Foundry SDK. In the example, the client application uses authentication to connect to the endpoint for the model, submit a prompt, and display the response.
46
-
47
-
>[!NOTE]
48
-
>In order to use the SDK, you need to install the `azure-ai-projects` package. The package is the core Azure AI Projects (Foundry) SDK used to connect to your Foundry project and obtain an OpenAI-compatible client.
45
+
After you created a **Foundry project** and **deployed a chat model** (for example, `gpt-4.1`), you can use the Foundry SDK. In the example, the client application uses authentication to connect to the endpoint for the model, submit a prompt, and display the response.
49
46
50
47
```python
51
48
# pip install openai>=1.3.0
@@ -77,11 +74,12 @@ In Microsoft Foundry, **generative AI models** and **agents** are related but se
77
74
-**Models = raw intelligence**
78
75
-**Agents = packaged, task‑oriented workers built on top of that intelligence**
79
76
80
-
When you use a generative AI model on its own:
77
+
When you use a generative AI model on its own:
78
+
81
79
- You want pure inference: "Take this prompt and generate output."
82
80
- You’re experimenting in the Playground
83
81
- You call the model via the **OpenAI Responses API**
0 commit comments