
LANGCHAIN — Hackathon Kickoff and Project Ideas
Social media is not about the exploitation of technology but service to community. — Simon Mainwaring
Streamlit’s LLM Hackathon is a great opportunity to showcase your skills and build innovative language models. In this article, we’ll explore some resources and project ideas for the hackathon. We’ll also delve into examples of LangChain and Streamlit integration and discuss the LangChain Prompt Hub.
Getting Started with LangChain and Streamlit
To begin, let’s take a look at some reference implementations of LangChain agents as Streamlit apps. These include a search-enabled chatbot, a chat app with user feedback, a chatbot capable of answering queries by referring to custom documents, and a chatbot that can communicate with your database. These implementations are available in the LangChain GitHub repository.
Additionally, you can explore a LangSmith cookbook on building a Streamlit Chat UI with LangSmith. This tutorial provides step-by-step instructions on creating a ChatGPT-like web app in Streamlit, supporting streaming, custom instructions, app feedback, and more.
For further inspiration, you can check out some LLM app ideas from the Streamlit blog, such as building a data analysis chatbot, generating data-driven presentations, and creating an app that lets you chat with “Warren Buffet.”
LangChain Prompt Hub
The LangChain Prompt Hub is a platform for submitting, discovering, inspecting, and remixing prompts. This tool is designed to help developers build faster and to collectively enhance the community’s expertise in prompting. You can contribute prompts and try out prompts submitted by others. The hub’s documentation provides a quickstart guide for getting involved.
Projects We’d Love to See
As for the hackathon projects, LangChain is interested in applications that utilize open source models, innovative retrieval tactics, practical agents, adventurous agents, and apps that connect unique data sources or formats.
Conclusion
The LLM Hackathon is an exciting opportunity to showcase your creativity and skills in building language models. With the resources and ideas presented here, you can kickstart your project and contribute to the LangChain community. We can’t wait to see what you build!
Remember to share your progress and projects with LangChain on Twitter, via email, or in the Streamlit Discord. Happy hacking!
