Skip to content

microsoft/RAG_Hack

Repository files navigation

RAGHack: Let's build RAG applications together

RAG AI Hack Banner

🛠️ Build, innovate, and #Hacktogether! 🛠️ It's time to start building AI applications using the power of RAG (Retrieval Augmented Generation). 🤖 + 📚 = 🔥

Large language models are powerful language generators, but they don't know everything about the world. RAG (Retrieval Augmented Generation) combines the power of large language models with the knowledge of a search engine. This allows you to ask questions of your own data, and get answers that are relevant to the context of your question.

RAGHack is your opportunity to get deep into RAG and start building RAG yourself. Across 25+ live streams, we'll show you how to build RAG apps on top of Azure AI in multiple languages (Python, Java, JS, C#) with multiple retrievers (AI Search, PostgreSQL, Azure SQL, Cosmos DB), with your own data sources! You'll learn about the most popular frameworks, like LangChain and Semantic Kernel, plus the latest technology, like agents and vision models. The possibilities are endless for what you can create... plus you can submit your hack for a chance to win exciting prizes! 🥳

The streams start September 3rd and end September 13th. Hack submissions are due September 16th, 11:59 PM PST. Join us!

Register

Register for the hackathon using any of the sessions linked on the Reactor series home page. This will register you for both the selected session and the hackathon.

Introduce yourself and look for teammates here GitHub Discussions!

Official Rules

Read the official rules 📃

Stream Schedule

English

Week 1: September 3rd-6th

Day/Time Topic Resources
9/3, 04:30 PM UTC / 09:30 AM PT RAG 101 Link
9/3, 06:00 PM UTC / 11:00 AM PT RAG with .NET Link
9/3, 08:00 PM UTC / 01:00 PM PT RAG with Azure AI Studio Link
9/3, 10:00 PM UTC / 03:00 PM PT RAG with Python Link
9/4, 03:00 PM UTC / 08:00 AM PT RAG with Langchain4J Link
9/4, 03:00 PM UTC / 08:00 AM PT RAG with LangchainJS Link
9/4, 09:00 PM UTC / 02:00 PM PT Scalable RAG with CosmosDB for NoSQL Link
9/5, 03:00 PM UTC / 08:00 AM PT Responsible AI Link
9/5, 05:00 PM UTC / 10:00 AM PT RAG on Cosmos DB MongoDB Link
9/5, 07:00 PM UTC / 12:00 PM PT RAG with Azure AI Search Link
9/5, 09:00 PM UTC / 02:00 PM PT RAG on PostgreSQL Link
9/5, 11:00 PM UTC / 04:00 PM PT RAG on Azure SQL Server Link
9/6, 04:00 PM UTC / 09:00 AM PT Intro to GraphRAG Link
9/6, 06:00 PM UTC / 11:00 AM PT Add multi-channel communication in RAG apps Link

Week 2: September 9th-13th

Day/Time Topic Resources
9/9, 03:00 PM UTC / 08:00 AM PT RAG with Java + Semantic Kernel Link
9/9, 05:00 PM UTC / 10:00 AM PT RAG with Java + Spring AI Link
9/9, 08:00 PM UTC / 01:00 PM PT RAG with vision models Link
9/9, 11:00 PM UTC / 04:00 PM PT Internationalization for RAG apps Link
9/10, 03:00 PM UTC / 08:00 AM PT Use Phi-3 to create a VSCode chat agent extension Link
9/10, 05:00 PM UTC / 10:00 AM PT Agentic RAG with Langchain Link
9/10, 10:00 PM UTC / 03:00 PM PT Build an OpenAI code interpreter for Python Link
9/11, 03:00 PM UTC / 08:00 AM PT Connections in Azure AI Studio Link
9/11, 05:00 PM UTC / 10:00 AM PT Explore AutoGen concepts with AutoGen Studio Link
9/11, 08:00 PM UTC / 01:00 PM PT RAG with Data Access Control Link
9/11, 10:00 PM UTC / 03:00 PM PT RAFT: (RAG + Fine Tuning) in Azure AI Studio Link
9/12, 04:00 AM UTC / 09:00 AM PT Pick the right model for the right job Link
9/12, 08:00 PM UTC / 01:00 PM PT Evaluating your RAG Chat App Link

Spanish

Day/Time Topic Resources
9/3, 03:00 PM UTC / 08:00 AM PT RAG: Generación Aumentada de Recuperación Link
9/4, 03:00 PM UTC / 08:00 AM PT RAG: Prácticas recomendadas de Azure AI Search Link
9/11, 03:00 PM UTC / 08:00 AM PT AI Multi-Agentes: Patrones, Problemas y Soluciones Link

Portuguese

Day/Time Topic Resources
9/3, 03:00 PM UTC / 08:00 AM PT RAG (Geração Aumentada de Busca) no Azure Link
9/12, 03:00 PM UTC / 08:00 AM PT Construindo RAG com Azure AI Studio e Python Link
9/13, 03:00 PM UTC / 08:00 AM PT Implantando RAG com .NET e Azure Developer CLI Link

Chinese

Day/Time Topic Resources
9/3, 12:30 PM UTC / 05:30 AM PT Global RAG Hack Together Link
9/10, 12:30 PM UTC / 05:30 AM PT Create RAG apps with Azure AI SDK Link
9/12, 12:30 PM UTC / 05:30 AM PT Create RAG applications with AI Toolkit VSCode Extension Link
9/14, 12:30 PM UTC / 05:30 AM PT Intro to GraphRAG Link

Office Hours

For additional help with your hacks, you can drop by Office Hours in our AI Discord channel. Here are the Office Hours scheduled so far:

Day/Time Topic/Hosts
9/4, 07:00 PM UTC / 12:00 PM PT Python, AI Search, Postgres, with Pamela
9/6, 07:00 PM UTC / 12:00 PM PT .NET with Bruno

RAG Code Samples

Repository Language/Retriever Costs
azure-search-openai-demo Python, Azure AI Search Requires Azure deployment, follow guide for lower cost deployment
azure-search-openai-demo-java Java, Azure AI Search Requires Azure deployment, see cost estimate for App Service deployment, Container Apps, Kubernetes
serverless-chat-langchainjs JavaScript, CosmosDB Can be run locally for free with Ollama, see cost estimate for Azure deployment
azure-search-openai-demo-csharp C#, Azure AI Search Requires Azure deployment, see cost estimate or follow guide for low cost deployment
rag-postgres-openai-python Python, PostgreSQL Can be run locally for free with Ollama, see cost estimate for deployment for Azure deployment.
Cosmic-Food-RAG-app Python, Cosmos DB MongoDB Requires Azure deployment, see cost estimate
contoso-chat Python, Azure AI Search, Azure AI Studio, PromptFlow Requires Azure deployment, see cost estimate
azure-sql-db-session-recommender-v2 C#, Azure SQL Can be run locally for free with Azure SQL Database free tier

To find more samples, check out the following resources:

Recommended Learning Materials 📚

Project Submission

Hack submissions are due September 16th, 11:59 PM PST.

Submit your project here when it's ready: 🚀 Project Submission

Check out this video for step by step project submission guidance: Project Submission Video

Prizes and Categories 🏅

Projects will be evaluated by a panel of judges, including Microsoft engineers, product managers, and developer advocates. Judging criteria will include innovation, impact, technical usability, and alignment with corresponding hackathon category.

Each winning team in the categories below will receive a cash prize of $500. 💸

  • Best overall
  • Best in JavaScript/TypeScript
  • Best in Java
  • Best in .NET
  • Best in Python
  • Best use of AI Studio
  • Best use of AI Search
  • Best use of PostgreSQL
  • Best use of Cosmos DB
  • Best use of Azure SQL

All hackathon participants who submit a project will receive a digital badge (sometime in October).

About

Hack Together: RAG Hack | Register, Learn, Hack

Topics

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published