Open Hack: Innovating Apps with Azure AI Studio

This OpenHack is an integration between Azure and OpenAI's large language models. It leverages Azure AI Search for data retrieval and ChatGPT-style Q&A interactions. Using the Retrieval Augmented Generation (RAG) design pattern with Azure Open AI's GPT models, it provides a natural language interaction to discover relevant responses to user queries. Azure AI Search simplifies data ingestion, transformation, indexing, and multilingual translation. The OpenHack includes sample data so it's ready to try end to end.

About the Hack

The Innovating Apps with Azure OpenAI OpenHack will take you on a series of challenges to develop an end-to-end solution using Microsoft’s Generative AI offering. The OpenHack consists of a series of challenges, in which your team must use the guidance and resources provided to implement intelligent search solutions on Azure. These challenges will increase in complexity as you progress through the OpenHack, staring from the creation of a project in Azure AI Studio to the deployment of our desired solution into a web app.

The goal of this OpenHack is to provide you the ability to work as a team, learn from each other, with greater knowledge retention as each team will have to “figure out” how to reach the success criteria. The coaches provide technical guidance, but not answers to the squads. In the beginning of each challenge one of the coaches will be providing lectures, demos to setup challenges and review the previous challenge’s solutions.

Get to know more about the OpenHack format on What The Hack

Challenges

1. Challenge 01: Creation of the Project in Azure AI Studio
   • Create your Azure AI hub and AI Project to start working on your solution
2. Challenge 02: Deploy and test a chat model without your data
   • Create your Azure OpenAI deployments and test your chat without data
3. Challenge 03: Add your Data to your project
   • Let's start working with the data of Contoso! Let's get to know more about Contoso's product offering
4. Challenge 04: Orchestrate and customize the set up with Prompt Flow
   • Now the real RAG begins: integrate the several components (datasources, ai search, etc.) in a logical flow
5. Challenge 05: Evaluate the chat performance
   • How good is your chat to answer the typical questions from your customers?
6. Challenge 06: Deploy your WebApp
   • The app is ready to get tested: let's deploy our app into a web application
7. Challenge 07: LLMOps in Visual Studio Code
   • Let's start working on your knowledge of LLMOps on Github!

Contributors

• Marta Santos
• Renato Ribeiro