E-SAS ChatBot

Introduction

Today, Large Language Models (LLMs) are being used more in different bussiness. However, fine-tunning is not the most suitable option when the LLM has to deal with private information.

LangChain is a tool which enables to connect data from several type of unstructured documents that contains text information. LangChain allows to retrieve the information from this data.

The goal of this work was to create a telegram chat bot which runs a LLM with langchain to access data from txt files which contains information of the bussiness.

Data obtained:

The data was created for the purpose of this project. In this case, it was created a fake company called E-SAS, E-SAS aims to sell pet accessories online.

Importance of the project:

Having a LLM which goes beyond the data trained from the typical ChatGPT model.

Tools used:

LangChain
Python
OpeanAI API
Telegram bot

Workflow of the project:

The data was loaded, cleaned, and saved as Document type.
Texts were transformed into embeddings.
Information retrieved and send to the LLM.
Telegram sends the answer.
The following flowchart shows how this process works out.

Show code

Follow me on my social media channels to know more about my projects.

Get In Touch

Pujilí, Cotopaxi, Ecuador

sebitas.alejo@hotmail.com

Quick Links

Home About Us Our Services Contact Us