This folder contains the public facing files for the lab WRK550.
These instructions are for participants of the Workshop "Build a Retail Copilot Code-First on Azure AI" at Microsoft AI Tour 2024-2025. The workshop is offered as an instructor-led session (WRK550) on the Prototype to Production track:
In this workshop, attendees will be introduced to the concept of GenAIOps, and learn to build a custom RAG-based retail copilot end-to-end, code-first on the Azure AI platform.
Using the provided sample, attendees will learn to prototype, evaluate and deploy a customer support chat AI for Contoso Outdoors - a fictitious retailer who sells hiking and camping equipment online. The sample uses a Retrieval Augmented Generation (RAG) architecture to implement a retail copilot solution that responds to customer queries with answers grounded in the retailer's product catalog and customer purchase history.
🌟 | The Microsoft AI Tour is Live and coming to a city near you!
- Register to attend at a tour stop near you.
- View Lab resources to continue your journey.
If you're not a workshop participant at AI Tour, you can still visit Azure-Samples/contoso-chat to explore the workshop using your own Azure subscription.
- View the workshop instructions in the
docs/workshopfolder - View a web preview of instructions at https://aka.ms/aitour/contoso-chat/workshop/
To participate in the instructor-led workshop you will need the following:
- Your own laptop. - It need only be capable of running a browser and GitHub Codespaces. We recommend using a recent version of the Edge, Chrome or Safari browser.
- A GitHub Account. - You can sign up for a free account if needed. Once you complete the workshop, you will have a personal fork of the sample (with workshop instructions) to continue exploring at home.
- recommended: Familiarity with Visual Studio Code. - We will run all code in GitHub Codespaces and use VS Code Online as our primary development environment in it.
- preferred: Familiarity with the
bashshell. - We'll use it in the VS Code terminal for command-line execution of tasks and scripts, including for Azure CLI commands. - preferred: Familiarity with Python and Jupyter Notebooks - We'll be creating Python scripts and running them from the command line and from Notebooks.
Important
If you are working on this at home, using the Self-Guided option, you also need:
- Active Azure Subscription - with access to the relevant resources and AI model deployments required for this workshop.
- Manual Provisioning Step - you will need to provision the infrastructure yourself by manually running required commands (
azd up).
To get started, you will need to:
- launch the Skillable Lab (in the Microsoft AI Tour instructor-led session) to get access to the relevant Azure subscription credentials.
- fork the contoso-chat sample (with branches)
- switch to the
aitour-WRK550branch & launch GitHub Codespaces - then follow the workshop instructions as described below
The workshop instructions are available in the repository under the docs/workshop folder as readable Markdown files. The content is authored in Markdown and is readable directly from source. However, for convenience, you can use one of these two options to get a rendered preview as shown below, which may be easier to navigate.
Visit https://aka.ms/aitour/contoso-chat/workshop in your browser:
- get an interactive hosted preview of this workshop guide you can explore now.
- bookmark and revisit it (we will keep it updated for the duration of the tour)
Launch GitHub Codespaces on your fork of the repo, then get the local preview by:
- changing to the workshop directory in a terminal:
cd docs/workshop - running the preview server:
mkdocs serve -a localhost:5000
You will see a dialog pop-up with options to view this in browser or in VS Code.
You can now continue working through the instructions for the workshop, starting with the Tour Guide Setup section. The figure below provides a high-level view of the developer workflow you will follow in this workshop.
Tip
If you enjoyed this workshop, please give the repository a ⭐, watch it for updates, and share it with others so we can help make this better for everyone!
Contoso Chat is an open source project supported by Microsoft. See the SUPPORT.md file for details on how to raise issues or contribute.
| Resources | Links | Description |
|---|---|---|
| Open-Source Samples | Contoso Chat · Contoso Web | Open-source repos with chat AI (backend) and web UI (frontend) samples used in the workshop. |
| Prompty | Docs · Extension · Package | New asset class that brings flexibility, observability, portability and understandability to the prompt engineering and ideation phase of app development. Currently in preview. |
| Azure AI Templates | Curated Collection | Curated collection of Azure AI azd templates that provide open-source samples for signature application scenarios using Azure AI and popular frameworks. Deployable with a single command! |
| Azure AI Foundry | Docs · SDK · Model Catalog | Unified platform for building end-to-end generative AI applications on Azure. Explore the model catalog, deploy and manage models and applications, adopt responsible AI practices. |
| Generative AI For Beginners | Open-Source Curriculum | Beginner-friendly open-source collection of lessons with hands-on projects to build up your knowledge of core concepts in prompt engineering, fine tuning, and more. |
The source code for this session can be found in the Contoso-Chat repo.