https://github.com/nitya/aistudio-rag-copilot-sandbox

Project to explore the Azure AI Studio RAG-based Copilot tutorials and documentation - with the goal of having a sandbox to explore related tools, techniques and best practices
https://github.com/nitya/aistudio-rag-copilot-sandbox

azure azure-ai-studio copilot llmops openai rag retail workshop

Last synced: 4 months ago
JSON representation

Project to explore the Azure AI Studio RAG-based Copilot tutorials and documentation - with the goal of having a sandbox to explore related tools, techniques and best practices

• Host: GitHub
• URL: https://github.com/nitya/aistudio-rag-copilot-sandbox
• Owner: nitya
• License: mit
• Created: 2024-07-30T13:48:42.000Z (10 months ago)
• Default Branch: main
• Last Pushed: 2024-08-12T12:46:27.000Z (10 months ago)
• Last Synced: 2024-08-12T16:36:54.042Z (10 months ago)
• Topics: azure, azure-ai-studio, copilot, llmops, openai, rag, retail, workshop
• Language: Dockerfile
• Homepage: https://nitya.github.io/aistudio-rag-copilot-sandbox/
• Size: 3.19 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Build a RAG-based Custom Copilot Code-First on Azure AI 

## Introducing Contoso Chat

[Contoso Chat](https://aka.ms/aitour/contoso-chat) is an open-source reference sample that implements an end-to-end solution for a **retail copilot** based on your data. The reference sample showcases the following tools and features:

 - Application **design** with [Prompty](https://prompty.ai) templates.

 - Application **development** with [Promptflow](https://aka.ms/promptflow) flex flows.

 - Application **evaluation** with [Azure AI Studio](https://aka.ms/promptflow) tooling

 - Application **provisioning** and **deployment**  with [Azure Developer CLI](https://aka.ms/azd)

The sample comes with workshop-style instructions to **deconstruct** the end-to-end development workflow, and get hands-on experience with how a custom copilot solution is built **code-first** with Azure AI tools and services.

## Reconstructing Contoso Chat

This repository focuses on **reconstructing** that experience from scratch, building a custom copilot with your data step-by-step, using the following resources. Along the way, we'll get more familiar with core concepts for generative AI, core developer tools for Azure AI, and the LLMOps end-to-end workflow from prompt design to production deployment.

- [ ] [Quickstart: Create Azure AI Hub & Project from UI](https://learn.microsoft.com/en-us/azure/ai-studio/quickstarts/get-started-playground) - **Tutorial**

- [ ] [Quickstart: Create Azure AI Hub & Project from SDK](https://learn.microsoft.com/en-us/azure/ai-studio/how-to/develop/create-hub-project-sdk?tabs=python) - **Tutorial**

- [ ] [Build a RAG-based copilot with promptflow SDK](https://learn.microsoft.com/en-us/azure/ai-studio/tutorials/copilot-sdk-build-rag?tabs=azure-portal) - **Tutorial: Part 1**

- [ ] [Evaluate and deploy your copilot with promptflow SDK](https://learn.microsoft.com/en-us/azure/ai-studio/tutorials/copilot-sdk-evaluate-deploy) - **Tutorial: Part 2**

- [ ] [Build a custom chat app using promptflow SDK](https://learn.microsoft.com/en-us/azure/ai-studio/quickstarts/get-started-code?tabs=macos) - **Tutorial**

- [ ] [Copilot application that implements RAG](https://github.com/Azure-Samples/rag-data-openai-python-promptflow) - **Azure Samples**

- [ ] [Contoso Chat Retail Copilot Application](https://github.com/Azure-Samples/contoso-chat) - **Azure Samples**

Let's get started.

---

## Setup Environment

Updating this codebase to use the same dev container as [Contoso Chat](https://github.com/Azure-Samples/contoso-chat/tree/main/.devcontainer) -