ragtime

An experimental RAG application.

1. Python backend, loading, querying data into a vector store, providing a HTTP API
2. A React frontend to interact with the agent querying the data

Installation

Make sure you have poetry and Bun installed!

poetry install

cd frontend

bun install

Running

cp .env.example .env
<Add your API keys to .env>

docker compose up
poetry run ragtime --serve &
cd frontend
bun vite dev

Data loading

To load data, find a youtube channel or playlist, for example Life Lessons, Tips, and Advice from Tim Ferriss View full playlist.

Then you can run:

poetry run yt-tools https://www.youtube.com/playlist?list=PLuu6fDad2eJyj3ZHfm9TlWWUNmqhdN2iZ

This will create a directory based on the channel name, eg Tim Ferriss and save all video's transcripts and metadata as JSON files in this folder.

Then you can load the data with: (Make sure you ran docker compose up)

poetry run ragtime --load "./Tim Ferriss"