From Basic to Custom: AI Chatbot Building 101

AI Custom Chatbot Quickstart

To get started leveling up your custom AI chatbot, check out our GitHub repository here.

Domain specific AI chatbots are powerful applications that can be used in many use cases:

• Virtual Assistants
• Knowledge retrieval
• Text synthesis
• Text formatting
• Sentiment analysis

Building AI chatbot apps is easy. Customizing is hard.

This ready-to-go sample project shows you how to build a custom chatbot using our preferred LLM tech stack:

• OpenAI
• Langchain
• Llama Hub Data Loaders
• Pinecone Vector Database
• Python (FastAPI web framework)

We are leveraging Retrieval Augmented Generation (RAG).

The goal of this tutorial is to demonstrate the basics.

• Demonstrate how to add custom data to an LLM model (we're using OpenAI's gpt-3.5-turbo)
• Demonstrate a conversational memory LLM chatbot
• Demonstrate using agents and tools with the Langchain Framework

Technical Objectives

• Ingest data into a vector database
• Query the vector database
• Query an agent that decides whether to query the vector database

Here’s a peak at the components of a basic AI custom chatbot.

For more resources: check out our other blog posts on AI.