Build Your First AI Chatbot

Hour 1: Pain Demo

# rule_based_bot.py - The OLD way

user_input = input("You: ").lower()

if "hello" in user_input:
    print("Bot: Hi there!")
elif "how are you" in user_input:
    print("Bot: I'm doing well!")
elif "weather" in user_input:
    print("Bot: I don't know the weather.")
else:
    print("Bot: I don't understand.")

Now try:

• "hey there!" - ❌ Doesn't work
• "what's up?" - ❌ Doesn't work
• "howdy!" - ❌ Doesn't work

Ask the student: "How many if statements would you need to handle every possible way someone might say hello? 100? 1000? It's infinite!"

This is why we need AI. Let's connect to a real LLM.

Hours 2-3: Guided Build

Step 1: Get Your API Key

1. Go to platform.openai.com
2. Sign up or log in
3. Go to API Keys section
4. Create a new secret key
5. Save it somewhere safe! You won't see it again.

Step 2: Install the OpenAI Package

pip install openai

Step 3: Set Up Your API Key

Create a file called .env:

OPENAI_API_KEY=sk-your-key-here

Or set it in your terminal:

# Windows
set OPENAI_API_KEY=sk-your-key-here

# Mac/Linux
export OPENAI_API_KEY=sk-your-key-here

Step 4: Your First API Call

# first_ai.py

from openai import OpenAI

# Create the client (it reads the API key automatically)
client = OpenAI()

# Make a request to the AI
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)

# Print the AI's response
print(response.choices[0].message.content)

Run it! You just talked to AI!

Step 5: Understanding Messages

The AI uses a conversation format with roles:

messages = [
    {"role": "user", "content": "What is Python?"},      # You
    {"role": "assistant", "content": "Python is..."},   # AI's response
    {"role": "user", "content": "How do I install it?"} # You again
]

Step 6: Build a Chat Loop

# chatbot.py

from openai import OpenAI

client = OpenAI()
messages = []  # Store conversation history

print("Chat with AI! Type 'quit' to exit.\n")

while True:
    # Get user input
    user_input = input("You: ")

    if user_input.lower() == 'quit':
        print("Goodbye!")
        break

    # Add user message to history
    messages.append({"role": "user", "content": user_input})

    # Get AI response
    response = client.chat.completions.create(
        model="gpt-4o-mini",
        messages=messages
    )

    # Extract the response text
    ai_message = response.choices[0].message.content

    # Add AI response to history
    messages.append({"role": "assistant", "content": ai_message})

    # Print it
    print(f"AI: {ai_message}\n")

Try it! Now say "hey there", "what's up", "howdy" - it understands everything!

Hours 4-5: Independent Build

Hours 6-7: Challenge Project

Requirements:

• Ask for the user's name when starting
• The AI should use the name naturally in responses
• If you tell it facts about yourself, it should remember them

Hint: You can "prime" the conversation:

name = input("What's your name? ")

messages = [
    {"role": "user", "content": f"My name is {name}. Please use my name occasionally when talking to me."},
    {"role": "assistant", "content": f"Nice to meet you, {name}! I'll remember that. What would you like to talk about?"}
]

# Now continue with the chat loop...

Hour 8: Review & Reflect