MongoDB GenAI Hackathon SF
Welcome to the MongoDB GenAI Hackathon! Below are all the resources you will need for the day.
MongoDB is renowned in the industry as a top-tier operational database. With MongoDB Atlas, we have unified MongoDB’s operational and vector database capabilities into a single unified platform. With Atlas, you can combine the benefits of a flexible document model and an extensive aggregation framework with vector search to build powerful AI applications for various use cases. As enterprises go from RAG to agents, MongoDB Atlas can also serve as long and short-term memory for agents.
Lightning Talk
Quickstarts
Credits
To get $100 Atlas credits, do the following:
- Sign-up for or login to your MongoDB Atlas account
- Enter the code provided on-site into the “Apply Code” box at the bottom of the MongoDB Atlas billing page.
Code Samples
- The GenAI-Showcase GitHub repo has several examples of using MongoDB Atlas with frameworks such as LangChain, and LlamaIndex as well as model providers such as OpenAI, Cohere, Hugging Face, etc.
- MongoDB Chatbot Framework is a set of libraries for building full-stack intelligent chatbot applications using MongoDB Atlas Vector Search.
Generating Embeddings
- Embeddings Generator is a simple tool to add embeddings to a dataset of your choice and ingest them into MongoDB.
RAG
Chatbots and Agents
Memory
Miscellaneous
Fireworks.ai provides a fast, reliable platform for developers to run generative artificial intelligence (AI) models at scale. Use Fireworks to provide your users with delightful, responsive experiences by delivering up to 4x lower latency than alternative solutions with zero compromise on model quality.
LlamaIndex is a simple, flexible data framework for connecting custom data sources to large language models (LLMs).
- LlamaIndex is free to use for Python and TypeScript. Our world-class document parsing API is LlamaParse, which is free for up to 1000 pages/day. You can sign up at cloud.llamaindex.ai.
Upstage provides a compact yet powerful Solar LLM with full-stack RAG components, including layout analysis, groundness checker, and embeddings. Use Upstage’s full-stack LLM components to provide a fast and efficient end-to-end LLM application.