Language Model Chatbot Project

Start with Why

Besides being an engineer, I am a Founding Member at No Pressure Improv - a performing arts group in Bangalore. As the name suggests, we improvise!

There is a lot of curiosity around improv. That means a lot of people want to know about improv! (Yay!) And also that people have to answer those questions a lot of time. (Oh...)

Enter AI

This is a prototype to answer commonly asked questions with AI. Like magic! This chatbot project leverages Google Gemini and LangChain, along with a custom dataset of frequently asked questions (FAQs).

Project Features

Chatbot Implementation:

• Uses Google Gemini and LangChain to create an intelligent and interactive chatbot.
• Trained on a custom dataset consisting of FAQs from my improv troupe to enhance contextual understanding.

FAQ Dataset:

• Curated a dataset specifically tailored to the unique queries and expressions commonly encountered within the improv community.
• The dataset enriches the chatbot's ability to provide relevant and context-aware responses.

🚧 Work in Progress

The current focus of this project is to address dependency-related errors.

Demo/ How to see this Chatbot in Action

• This project is uploaded as a Colab Notebook.
• To run this notebook
  1. Copy the notebook into your Drive,
  2. Download the dataset from here
  3. Upload onto Colab's directory
  4. Run the cells

Voila!