RAG-a-thon Developer Resources

AI Makerspace Resources

Access valuable tools and materials for both the Pre-Event Workshop and Main Event

• Extensive library of Gen AI tutorials, notebooks, and best practices
• Build your first LLM with our step-by-step Github repo
• Learn everything you need to know about LLM Ops in this free, open-source cohort
• Join the largest community of builders in the "Build, Ship, Share! Discord channel

LLAMAINDEX RESOURCES

• Open source libraries in Python and TypeScript
• LlamaParse, our industry-leading parser of complex document types
• LlamaExtract for structured data extraction
• llama-agents for creating multi-agent systems
• LlamaHub, our registry of open source data connectors, agent tools, and more 

PINECONE RESOURCES:

Getting started with Pinecone

• Our learn page, which has a ton of worked-out guides, handbooks, and blog posts on using Pinecone and AI tooling: https://www.pinecone.io/learn/
• Our quickstart for using Pinecone: https://docs.pinecone.io/guides/get-started/quickstart
• Our sample app page, which has working instances of RAG, Assistant, and semantic search for Next.js devs: https://docs.pinecone.io/examples/sample-apps
• Our examples repo where a plethora of notebooks exist, ready to use with Colab! https://github.com/pinecone-io/examples

Pinecone Inference (Embeddings and Reranker)
In addition to our vector database, we have our Pinecone Inference offering which allows users to embed data and rerank embeddings in the same endpoint:

• Docs: https://docs.pinecone.io/guides/inference/understanding-inference
• Arjun Patel's article explaining how to use Inference embeddings with a worked out example: https://www.pinecone.io/learn/the-practitioners-guide-to-e5/
  
• Reranker docs: https://www.pinecone.io/learn/refine-with-rerank/

Pinecone Assistant
Finally, we have our assistants offering which lets you create a chatbot by uploading pdfs directly to Pinecone. 
Docs: https://docs.pinecone.io/guides/assistant/understanding-assistant

VESSL AI RESOURCES:

• Agentic RAG-A-THON VESSL AI Resources

Additionally, here are some helpful links to other key resources:

• VESSL Hub: Where users can launch pre-built examples and models.
• Example Code: Demo app code using LlamaIndex and Pinecone.
• LLM Deployment Guide: Instructions for deploying LLM APIs on VESSL.

ARIZE RESOURCES:

• Phoenix web portal- create an account here to access an online instance of Phoenix
• Phoenix documentation
• Open source libraries:
• Python Tracing & Evaluation - our main library, use this to trace, evaluate, and experiment on LLM Apps
• JS / TS Tracing - use this library to trace JS / TS apps
• Tracing, evaluation, and experimentation example notebooks - a set of examples that show Phoenix in action

BOX RESOURCES

In order to integrate with Box, you need a few things:

• A Box instance — if you are not a current Box customer, sign up for a free dev account.

• A Box app — more on how to create an app

• Your app approved in your Box instance — This is done by your admin. The good news is if you are using a free developer account, you are the admin.

TOOLHOUSE RESOURCES

• https://join.toolhouse.ai - This is where hackathon attendees can get our special offer ($1 for $150 worth of Toolhouse credits). We can give additional credits to attendees who are building on top of Toolhouse during the hackathon.
• https://discord.toolhouse.ai - Our Discord server
• https://docs.toolhouse.ai - Our documentation

Additional resources:

• Developer documentation

• Developer community

• Llama-Index readers

• Llama-Index agent tools

• Box Llama-Index blogs

Participants are welcome to use any open-source tool available.

Do you want to find other open-sourced models?

• Llama Stack GitHub Repository
  Access the core components and API specifications of the Llama Stack.
  GitHub - Llama Stack 
  
  
• Go to https://huggingface.co/ and explore models and datasets available at the top-right and hack-away. Also, their chat-ui is available at https://github.com/huggingface/chat-ui , Check out for inspiration.
  

Other:

Open AI 

• Sign up using your email and go to  https://platform.openai.com/account/api-keys
• Navigate to User -> API Keys, and Create a Secret key. Note: You can create multiple secret keys if necessary.