This is a pilot project to provide support to cult users via a chatbot. This is a LLM RAG Chatbot based on custom knowledge base.
Currently, only the "cult" and "gym" section faq answers are considered as the knowledge base.
# In the project repository
$ python3 -m venv venv
$ source venv/bin/activate
$ python -m pip install -r requirements.txt
$ python src/retriever/create_retriever.py
$ python invoke.py
| Key | Value Information | Required |
|---|---|---|
GOOGLE_API_KEY |
Required to make calls and connect to the GEMINI API, you can sign up for a client id here. | Yes |
TAVILY_API_KEY |
Required to make searches on the internet to run the agent, you can sign up for a client secret here. | Yes (Not if using OpenAI) |
LLM_MODEL |
The LLM model version that will be used across the codebase. | Yes |
EMBEDDING_MODEL |
The embedding model version that will be used across the codebase. | Yes |
Some additional details if they are required.
#General Queries about Cult, Programs offered, Gym etc.
-User will be provided the response based on our knowledge base.
#Specific Pricing Queries
- User will be provided a redirect link based on the query.
#Specific Gym Queries
- User will be provided a redirect link based on the query.
# For any question that is out-of-scope, a default answer will be provided.
