Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified 03-Azure/01-02 Data/03-Talk_to_your_data/Images/image005.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified 03-Azure/01-02 Data/03-Talk_to_your_data/Images/image006.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified 03-Azure/01-02 Data/03-Talk_to_your_data/Images/image009.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified 03-Azure/01-02 Data/03-Talk_to_your_data/Images/image013.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified 03-Azure/01-02 Data/03-Talk_to_your_data/Images/image042.png
Binary file modified 03-Azure/01-02 Data/03-Talk_to_your_data/Images/image044.png
2 changes: 1 addition & 1 deletion 03-Azure/01-02 Data/03-Talk_to_your_data/README.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
![image](./Images/Preview.png)

# Microsoft Migrate & Modernize MicroHack Day - Ask Analyze Act - Talk to Your Data in the Era of AI
# Ask Analyze Act - Talk to Your Data in the Era of AI
- [**MicroHack introduction**](#MicroHack-introduction)
- [**MicroHack context**](#microhack-context)
- [**Objectives**](#objectives)
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -4,21 +4,23 @@

## Goal

The goal of this exercise is to build a unified, analytics‑ and AIready data foundation in Microsoft Fabric that enables governed reporting, natural language querying, and intelligent data interaction through a Data Agent.
The goal of this exercise is to build a report-ready and AI-ready semantic layer in Microsoft Fabric by creating and optimizing a Semantic Model, refining a Power BI report, and preparing model metadata for natural language analysis with Copilot.


## Actions

* Combine mirrored Azure SQL Managed Instance databases and external CSV files into a single Lakehouse using shortcuts
* Create and optimize a Semantic Model from the Lakehouse tables, including relationships and time‑based logic
* Prepare the data for AI by simplifying the schema and providing AI instructions for business context
* Set up a Data Agent using the prepared Lakehouse and Semantic Model as trusted data sources
* Enable the required Power BI / Fabric trial to unlock reporting and AI capabilities
* Create and optimize a Semantic Model from the Lakehouse tables, including key relationships and cross-filter behavior
* Auto-create and refine a Power BI report using Copilot prompts and manual adjustments
* Prepare the semantic model for AI by simplifying the schema, reviewing verified answers, and adding AI instructions
* Explore Power BI Copilot with the prepared model to validate business-friendly responses

## Success criteria

* You have successfully unified operational and external data in a single Lakehouse that is accessible via Microsoft Fabric
* You have successfully created and optimized a Semantic Model that supports reliable reporting and efficient query performance
* You have successfully prepared the data schema and AI instructions to enable accurate natural language queries
* You have successfully validated that the Data Agent returns relevant, context‑aware answers based on trusted analytics data
* You have successfully enabled the Power BI / Fabric trial and can access reporting and Copilot features
* You have successfully created and optimized a Semantic Model that supports reliable reporting and efficient analysis
* You have successfully generated and refined a report that reflects key business insights
* You have successfully prepared the model for AI by simplifying schema exposure and adding business-focused AI instructions
* You have successfully validated Power BI Copilot responses against the prepared semantic model

[Open the step-by-step solution for Challenge 2](../walkthrough/challenge-02/solution-02.md)
22 changes: 12 additions & 10 deletions 03-Azure/01-02 Data/03-Talk_to_your_data/challenges/challenge-03.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,23 +4,25 @@

## Goal

The goal of this exercise is to understand how data context and instructions influence the behavior, reasoning quality, and trustworthiness of a Data Agent, and validate how a well‑configured agent performs in a real M365 Copilot user experience.
The goal of this exercise is to understand how instruction quality and data context influence Data Agent response quality, then validate those improvements from baseline behavior through production-ready guidance and M365 Copilot publishing.


## Actions

* Interact with the Data Agent using predefined prompts without any instructions to observe default behavior
* Design and apply custom Data Agent instructions
* Test the Data Agent again and evaluate response improvements
* Compare custom instructions with the provided Agent instructions and Data Source instructions
* Publish the Data Agent and access it through M365 Copilot chat
* Set up a Data Agent connected to the Lakehouse as the trusted data source
* Use prompt engineering principles and run prompts without instructions to observe baseline behavior
* Write your own Agent Instructions and Data Source Description/Instructions
* Compare your setup with the lab reference instructions across stepwise maturity levels (Step 0 to Step 4)
* Re-test the Data Agent to confirm improvements in clarity, structure, and reliability
* Publish the Data Agent and open it in M365 Copilot chat


## Success criteria

* You have successfully observed clear differences between uninstructed and instructed agent behavior
* You have successfully demonstrated that custom instructions improve the accuracy, clarity, and usefulness of agent responses
* You have successfully understood the impact of Agent instructions versus Data Source instructions
* You have successfully validated that the Data Agent behaves consistently when accessed via M365 Copilot
* You have successfully observed clear differences between baseline (no instructions) and instructed behavior
* You have successfully demonstrated that stronger instructions improve response quality step by step
* You have successfully understood the distinct impact of Agent Instructions versus Data Source Description/Instructions
* You have successfully validated improved Data Agent behavior with repeated prompt testing
* You have successfully published the agent and verified it in M365 Copilot

[Open the step-by-step solution for Challenge 3](../walkthrough/challenge-03/solution-03.md)
Loading
Loading