/bricky

Haystack/OpenAI based chatbot curating a custom knowledgebase

Primary LanguagePythonMIT LicenseMIT

Meet Bricky - a conversational bot using OpenAI 🤖

Remember clippy? Meet bricky!

Bricky is a conversational bot using Retrieval-Augmented Generation with some help from OpenAI's GPT-3 LLM.

Bricky indexes content stored in markdown files and vectorizes it using OpenAI embeddings. It then uses few-shot learning using a ChatGPT prompt to generate an answer based on relevant content.

Read more about my journey into this field and the background for creating Bricky in my blog article

The project is inspired by the awesome HoustonAI by Astro

screenshot.png

Getting started 🚀

Prereqs

Provide these env variables for the api container by creating a dotenv file in api/.env

OPENAI_KEY=<YOUR OPENAI KEY GOES HERE>

Steps

  1. Clone this repo!
  2. Copy over your documentation to api/sources
  3. Run docker-compose: docker-compose up

You should now have two endpoints running:

Note: if you make changes to the any files, i.e. api/.env or the docs in sources/docs, you need to rebuild the images: docker-compose rebuild --no-cache.

Learn more

To learn more about Haystack and OpenAI, take a look at the following resources:

To learn more about Next.js, take a look at the following resources:

Powered by haystack and OpenAI ChatGPT

  • Frontend implementation can be found here.
  • Backend implementation can be found here.

Questions or comments? Reach out to @larsbaunwall

Don't forget to ⭐ this repo!