Chatbot with Conversation Memory

LLMs have no built-in memory : they are stateless. Every time you send a message, the entire conversation history is sent to the API as a list of messages. The model reads all of it from scratch and replies in context. What feels like memory is just that growing list being passed on every call.

Features

• Conversation memory : the bot remembers everything you said in the session
• Streaming responses : replies appear word by word in real time
• Custom personas : give the bot any personality via the system prompt
• Temperature control : adjust how creative or focused the responses are
• Smart trimming : control how many messages the bot remembers
• Clear history : start fresh anytime with one click

---

Getting Started

1. Clone the repo

git clone https://github.com/YOUR_USERNAME/chatbot-memory.git
cd chatbot-memory

2. Create a virtual environment

python -m venv venv
source venv/bin/activate       # Mac/Linux
venv\Scripts\activate          # Windows

3. Install dependencies

pip install -r requirements.txt

4. Get a free Groq API key

1. Go to console.groq.com
2. Sign up and create an API key
3. No credit card needed

5. Create a .env file

GROQ_API_KEY=gsk_...

6. Run the app

streamlit run app.py

The app opens at http://localhost:8501 in your browser.

---

How to Use

Setting	What it does
System prompt	Give the bot a role e.g. "You are a Socratic tutor"
Temperature	0 = focused and precise, 1 = creative and unpredictable
Max messages	How many messages the bot remembers before forgetting older ones
Clear history	Wipe the conversation and start fresh

---

Acknowledgements

Built as part of a 7-day learning project to understand how LLMs, context windows, and API calls actually work.