Skip to content

Commit c11c1a0

Browse files
JamesJBarnettJamesJBarnett
authored
Merge pull request #50040 from GraemeMalcolm/main
Updates to AI modules
2 parents 85748f1 + cbc0357 commit c11c1a0

39 files changed

+799
-618
lines changed

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/1-introduction.yml

Lines changed: 16 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,16 @@
1-
### YamlMime:ModuleUnit
2-
uid: learn.wwl.build-copilot-ai-studio.introduction
3-
title: Introduction
4-
metadata:
5-
title: Introduction
6-
description: Explore building custom generative AI apps with Azure AI Foundry.
7-
author: madiepev
8-
ms.author: madiepev
9-
ms.date: 11/26/2024
10-
ms.topic: unit
11-
ms.collection:
12-
- wwl-ai-copilot
13-
durationInMinutes: 2
14-
content: |
15-
[!include[](includes/1-introduction.md)]
16-
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.introduction
3+
title: Introduction
4+
metadata:
5+
title: Introduction
6+
description: Explore building custom generative AI apps with Azure AI Foundry.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
durationInMinutes: 2
14+
content: |
15+
[!include[](includes/1-introduction.md)]
16+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/2-ground-language-model.yml

Lines changed: 16 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,16 @@
1-
### YamlMime:ModuleUnit
2-
uid: learn.wwl.build-copilot-ai-studio.ground-language-model
3-
title: Understand how to ground your language model
4-
metadata:
5-
title: Understand how to ground your language model
6-
description: Learn about the grounding language models to get factually accurate responses.
7-
author: madiepev
8-
ms.author: madiepev
9-
ms.date: 11/26/2024
10-
ms.topic: unit
11-
ms.collection:
12-
- wwl-ai-copilot
13-
durationInMinutes: 6
14-
content: |
15-
[!include[](includes/2-ground-language-model.md)]
16-
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.ground-language-model
3+
title: Understand how to ground your language model
4+
metadata:
5+
title: Understand how to ground your language model
6+
description: Learn about the grounding language models to get factually accurate responses.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
durationInMinutes: 6
14+
content: |
15+
[!include[](includes/2-ground-language-model.md)]
16+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/3-search-data.yml

Lines changed: 16 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,16 @@
1-
### YamlMime:ModuleUnit
2-
uid: learn.wwl.build-copilot-ai-studio.search-data
3-
title: Make your data searchable
4-
metadata:
5-
title: Make your data searchable
6-
description: Learn how to use Azure AI Search to make your data searchable and create an index.
7-
author: madiepev
8-
ms.author: madiepev
9-
ms.date: 11/26/2024
10-
ms.topic: unit
11-
ms.collection:
12-
- wwl-ai-copilot
13-
durationInMinutes: 7
14-
content: |
15-
[!include[](includes/3-search-data.md)]
16-
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.search-data
3+
title: Make your data searchable
4+
metadata:
5+
title: Make your data searchable
6+
description: Learn how to use Azure AI Search to make your data searchable and create an index.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
durationInMinutes: 7
14+
content: |
15+
[!include[](includes/3-search-data.md)]
16+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/3b-openai-client.yml

Lines changed: 17 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,17 @@
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.openai-client
3+
title: Create a RAG-based client application
4+
metadata:
5+
title: Create a RAG-based client application
6+
description: Learn how to use the Azure OpenAI SDK to create a RAG client.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
zone_pivot_groups: dev-lang-csharp-python
14+
durationInMinutes: 7
15+
content: |
16+
[!include[](includes/3b-openai-client.md)]
17+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/4-build-copilot.yml

Lines changed: 16 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,16 @@
1-
### YamlMime:ModuleUnit
2-
uid: learn.wwl.build-copilot-ai-studio.build-copilot
3-
title: Build an agent with prompt flow
4-
metadata:
5-
title: Build an agent with prompt flow
6-
description: Learn about build an agent with prompt flow.
7-
author: madiepev
8-
ms.author: madiepev
9-
ms.date: 11/26/2024
10-
ms.topic: unit
11-
ms.collection:
12-
- wwl-ai-copilot
13-
durationInMinutes: 7
14-
content: |
15-
[!include[](includes/4-build-copilot.md)]
16-
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.build-copilot
3+
title: Implement RAG in a prompt flow
4+
metadata:
5+
title: Implement RAG in a prompt flow
6+
description: Learn about using RAG in a prompt flow.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
durationInMinutes: 7
14+
content: |
15+
[!include[](includes/4-build-copilot.md)]
16+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/5-exercise.yml

Lines changed: 16 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,16 @@
1-
### YamlMime:ModuleUnit
2-
uid: learn.wwl.build-copilot-ai-studio.exercise
3-
title: Exercise - Create a custom agent that uses your own data
4-
metadata:
5-
title: Exercise - Create a custom agent that uses your own data
6-
description: Create a custom agent that uses your own data with the Azure AI Foundry portal.
7-
author: madiepev
8-
ms.author: madiepev
9-
ms.date: 11/26/2024
10-
ms.topic: unit
11-
ms.collection:
12-
- wwl-ai-copilot
13-
durationInMinutes: 15
14-
content: |
15-
[!include[](includes/5-exercise.md)]
16-
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.exercise
3+
title: Exercise - Create a generative AI app that uses your own data
4+
metadata:
5+
title: Exercise - Create a generative AI app that uses your own data
6+
description: Create a generative AI app that uses your own data with the Azure AI Foundry portal.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
durationInMinutes: 45
14+
content: |
15+
[!include[](includes/5-exercise.md)]
16+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/6-knowledge-check.yml

Lines changed: 50 additions & 50 deletions
Original file line numberDiff line numberDiff line change
@@ -1,50 +1,50 @@
1-
### YamlMime:ModuleUnit
2-
uid: learn.wwl.build-copilot-ai-studio.knowledge-check
3-
title: Module assessment
4-
metadata:
5-
title: Module assessment
6-
description: Knowledge check to test your knowledge on grounding models with RAG.
7-
author: madiepev
8-
ms.author: madiepev
9-
ms.date: 11/26/2024
10-
ms.topic: unit
11-
ms.collection:
12-
- wwl-ai-copilot
13-
durationInMinutes: 3
14-
content: |
15-
quiz:
16-
questions:
17-
- content: "What does groundedness refer to in the context of generative AI?"
18-
choices:
19-
- content: "The use of a locally deployed language model."
20-
isCorrect: false
21-
explanation: "Incorrect. RGroundedness doesn't refer to the location where the model is deployed."
22-
- content: "Using data to contextualize prompts and ensure relevant responses."
23-
isCorrect: true
24-
explanation: Correct. Groundedness involves using a data source to *ground* prompts and ensure accurate, relevant responses."
25-
- content: "Using the lowest possible number of tokens in a prompt."
26-
isCorrect: false
27-
explanation: "Incorrect. Groundedness does not minimize the number of tokens in a prompt."
28-
- content: "What pattern can you use to ground prompts?"
29-
choices:
30-
- content: "Metadata Optimized Prompt (MOP)."
31-
isCorrect: false
32-
explanation: "Incorrect. There is no MOP pattern for grounding prompts."
33-
- content: "Data Understanding Support Text (DUST)."
34-
isCorrect: false
35-
explanation: "Incorrect. There is no DUST pattern for grounding prompts."
36-
- content: "Retrieval Augmented Generation (RAG)."
37-
isCorrect: true
38-
explanation: "Correct. The RAG pattern is used to add grounding data to prompts."
39-
- content: "Which tool enables you to retrieve grounding data in a prompt flow?"
40-
choices:
41-
- content: "Query tool."
42-
isCorrect: false
43-
explanation: "Incorrect. There is no Query tool in Promot Flow."
44-
- content: "Content Safety tool."
45-
isCorrect: false
46-
explanation: "Incorrect. The Content Safety tool does not retrieve grounding data."
47-
- content: "Index Lookup tool."
48-
isCorrect: true
49-
explanation: "Correct. You can use the Index lookup tool to retrieve grounding data from an index in Prompt Flow."
50-
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.knowledge-check
3+
title: Module assessment
4+
metadata:
5+
title: Module assessment
6+
description: Knowledge check to test your knowledge on grounding models with RAG.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
durationInMinutes: 3
14+
content: |
15+
quiz:
16+
questions:
17+
- content: "What does groundedness refer to in the context of generative AI?"
18+
choices:
19+
- content: "The use of a locally deployed language model."
20+
isCorrect: false
21+
explanation: "Incorrect. RGroundedness doesn't refer to the location where the model is deployed."
22+
- content: "Using data to contextualize prompts and ensure relevant responses."
23+
isCorrect: true
24+
explanation: Correct. Groundedness involves using a data source to *ground* prompts and ensure accurate, relevant responses."
25+
- content: "Using the lowest possible number of tokens in a prompt."
26+
isCorrect: false
27+
explanation: "Incorrect. Groundedness doesn't minimize the number of tokens in a prompt."
28+
- content: "What pattern can you use to ground prompts?"
29+
choices:
30+
- content: "Metadata Optimized Prompt (MOP)."
31+
isCorrect: false
32+
explanation: "Incorrect. There's no MOP pattern for grounding prompts."
33+
- content: "Data Understanding Support Text (DUST)."
34+
isCorrect: false
35+
explanation: "Incorrect. There's no DUST pattern for grounding prompts."
36+
- content: "Retrieval Augmented Generation (RAG)."
37+
isCorrect: true
38+
explanation: "Correct. The RAG pattern is used to add grounding data to prompts."
39+
- content: "How can you use the RAG pattern in a client app that uses the Azure OpenAI SDK?"
40+
choices:
41+
- content: "Add text files containing the grounding data to the app folder."
42+
isCorrect: false
43+
explanation: "Incorrect. You shouldn't include grounding data as text files in the app folder."
44+
- content: "You don't need to do anything. Azure AI Foundry automatically grounds all prompts using Bing Search."
45+
isCorrect: false
46+
explanation: "Incorrect. Azure AI Foundry doesn't automatically ground all prompts using Bing Search."
47+
- content: "Add index connection details to the OpenAI ChatClient configuration."
48+
isCorrect: true
49+
explanation: "Correct. Adding index connection information to the client enables grounding of prompts."
50+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/7-summary.yml

Lines changed: 16 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,16 @@
1-
### YamlMime:ModuleUnit
2-
uid: learn.wwl.build-copilot-ai-studio.summary
3-
title: Summary
4-
metadata:
5-
title: Summary
6-
description: Summary of key learning points on building a custom agent with Azure AI Foundry.
7-
author: madiepev
8-
ms.author: madiepev
9-
ms.date: 11/26/2024
10-
ms.topic: unit
11-
ms.collection:
12-
- wwl-ai-copilot
13-
durationInMinutes: 1
14-
content: |
15-
[!include[](includes/7-summary.md)]
16-
1+
### YamlMime:ModuleUnit
2+
uid: learn.wwl.build-copilot-ai-studio.summary
3+
title: Summary
4+
metadata:
5+
title: Summary
6+
description: Summary of key learning points on building a custom agent with Azure AI Foundry.
7+
author: madiepev
8+
ms.author: madiepev
9+
ms.date: 04/17/2025
10+
ms.topic: unit
11+
ms.collection:
12+
- wwl-ai-copilot
13+
durationInMinutes: 1
14+
content: |
15+
[!include[](includes/7-summary.md)]
16+

‎learn-pr/wwl-data-ai/build-copilot-ai-studio/includes/1-introduction.md

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -4,7 +4,7 @@ One prevalent challenge when implementing language models through chat is the so
44

55
## Ungrounded prompts and responses
66

7-
When you use a language model to generate a response to a prompt, the only information that the model has to base the answer on comes from the data on which it was trained - which is often just large amounts of uncontextualized text from the Internet or some other source.
7+
When you use a language model to generate a response to a prompt, the only information that the model has to base the answer on comes from the data on which it was trained - which is often just a large volume of uncontextualized text from the Internet or some other source.
88

99
![Diagram of an ungrounded model returning an uncontextualized response.](../media/ungrounded.png)
1010

@@ -19,4 +19,3 @@ In contrast, you can use a data source to *ground* the prompt with some relevant
1919
The data source can be any repository of relevant data. For example, you could use data from a product catalog database to ground the prompt "Which product should I use to do *X*?" so that the response includes relevant details of products that exist in the catalog.
2020

2121
In this module, you explore how to create your own chat-based language model application that is grounded, by building an agent with your own data.
22-

0 commit comments

Comments
 (0)