Project: Custom Chatbot with Memory

In this project, you will build a custom chatbot that remembers the conversation history. It will use OpenAI's GPT API and a simple web interface with Streamlit.

A chatbot with memory can answer follow‑up questions correctly because it keeps track of previous messages.

What You Will Build

A web page where users type messages.
The chatbot responds using GPT‑3.5 or GPT‑4.
It remembers the conversation so you can ask follow‑ups like "What did I just ask?"

Prerequisites

Python installed (3.8+).
OpenAI API key (get it here).
Basic Python knowledge (functions, lists, dictionaries).

Step 1: Install Required Libraries

Open a terminal and run:

pip install openai streamlit

Step 2: Write the Chatbot Code

Create a file named chatbot.py and copy the code below. Replace your-api-key with your actual OpenAI API key.

import openai
import streamlit as st

# Set your OpenAI API key
openai.api_key = "your-api-key"

st.title("🤖 My Chatbot")

# Initialize session state to store conversation history
if "messages" not in st.session_state:
    st.session_state.messages = []

# Display previous messages
for msg in st.session_state.messages:
    with st.chat_message(msg["role"]):
        st.markdown(msg["content"])

# Get user input
if prompt := st.chat_input("Say something..."):
    # Add user message to history
    st.session_state.messages.append({"role": "user", "content": prompt})
    with st.chat_message("user"):
        st.markdown(prompt)

    # Call OpenAI API
    response = openai.ChatCompletion.create(
        model="gpt-3.5-turbo",
        messages=st.session_state.messages
    )
    reply = response.choices[0].message.content

    # Add assistant reply to history
    st.session_state.messages.append({"role": "assistant", "content": reply})
    with st.chat_message("assistant"):
        st.markdown(reply)

Step 3: Run the Chatbot

In the terminal, run:

streamlit run chatbot.py

Your browser will open with the chatbot interface. Type a message and press Enter. The bot will respond and remember the whole conversation.

How It Works

st.session_state.messages stores the entire chat history.
Each new user message is appended to the list.
The entire list is sent to OpenAI’s API – the model sees the whole conversation so far.
The assistant’s reply is also stored, enabling follow‑ups.

Try It Yourself

Ask: "What is the capital of France?" Then ask: "What about Germany?" The bot should understand you are still asking for capitals because it remembers the context.

Two Minute Drill

Streamlit provides an easy web interface.
Store conversation in st.session_state.
Send the full message history to OpenAI API each time.
The model uses the history to answer follow‑ups.

Need more clarification?

Drop us an email at career@quipoinfotech.com

Welcome to Quipoin

Quipoin Menu

Project: Custom Chatbot with Memory

Need more clarification?