🔌 1.02 – API Basics

📑 On this page

What is an API?
REST APIs
Authentication
Making Your First API Call
Rate Limits & Pricing
Exercises

🌐 What is an API?

📌 Definition

API (Application Programming Interface) is a set of rules that allows different software applications to communicate with each other. Think of it as a waiter in a restaurant – you give an order (request), the waiter takes it to the kitchen (API), and brings back your food (response).

💡 Why APIs matter for AI

When you use ChatGPT, you're actually using an API behind the scenes. Your prompts are sent to OpenAI's servers, processed by GPT-4, and the response comes back. You don't need to run the model on your computer!

📡 REST APIs

📌 REST (Representational State Transfer)

Most AI APIs use REST architecture. Key concepts:

Endpoint: A specific URL where the API can be accessed
HTTP Methods: GET (retrieve), POST (send), PUT (update), DELETE (remove)
Headers: Metadata about the request (authentication, content type)
Body: The actual data being sent (your prompt)

Method	Use with AI APIs	Example
GET	Retrieve model info, check usage	GET /v1/models
POST	Send prompts, get completions	POST /v1/chat/completions

🔐 Authentication

📌 API Keys

Most AI APIs require an API key – a unique secret string that identifies you and tracks your usage.

# Example API key header
Authorization: Bearer sk-xxxxxxxxxxxxxxxxxxxxxxxx

⚠️ Security Warning

Never share your API key
Never commit it to GitHub
Use environment variables instead of hardcoding
Rotate keys periodically

📞 Making Your First API Call

📌 OpenAI API Example (Chat Completion)

cURL (command line) example:

curl https://api.openai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gpt-4",
    "messages": [{"role": "user", "content": "What is an LLM?"}],
    "temperature": 0.7
  }'

Python example:

from openai import OpenAI

client = OpenAI(api_key="YOUR_API_KEY")

response = client.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": "What is an LLM?"}],
    temperature=0.7
)

print(response.choices[0].message.content)

JavaScript (Node.js) example:

import OpenAI from 'openai';

const openai = new OpenAI({ apiKey: 'YOUR_API_KEY' });

async function main() {
  const completion = await openai.chat.completions.create({
    model: 'gpt-4',
    messages: [{ role: 'user', content: 'What is an LLM?' }],
    temperature: 0.7,
  });
  console.log(completion.choices[0].message.content);
}
main();

💰 Rate Limits & Pricing

📌 Key Concepts

Rate limits: How many requests you can make per minute/day
Token-based pricing: Pay per token (input + output)
Different costs per model: GPT-4 is more expensive than GPT-3.5

Model	Input Price (per 1M tokens)	Output Price (per 1M tokens)
GPT-4 Turbo	$10.00	$30.00
GPT-3.5 Turbo	$0.50	$1.50
Claude 3 Sonnet	$3.00	$15.00
Claude 3 Haiku	$0.25	$1.25

✍️ Exercises

Exercise 1.02.1 – Understanding API responses

What would be the HTTP status code for each situation?

Your API call succeeds and returns data
Your API key is invalid
You hit your rate limit (too many requests)
The API endpoint doesn't exist

Answers:
1. 200 OK
2. 401 Unauthorized
3. 429 Too Many Requests
4. 404 Not Found

Exercise 1.02.2 – API key security

List three security best practices for handling API keys.

Answers:
1. Use environment variables (.env files) – never hardcode keys
2. Add your .env file to .gitignore
3. Rotate keys regularly and use different keys for development/production
4. Use API key restrictions (IP whitelisting or allowed endpoints when available)

Exercise 1.02.3 – Try it yourself

Setup Tips:
- OpenAI: platform.openai.com → Sign up → API keys → Create new key
- Start with $5 free credit (new users)
- Run the Python example above after installing `pip install openai`
- Save your key as an environment variable to keep it secure

📌 Key Takeaways

APIs allow applications to communicate – AI APIs let you call LLMs without running them locally
REST APIs use HTTP methods (GET, POST) and endpoints
Authentication requires API keys – keep them secure!
You pay per token – costs vary by model

📘 Next Module: 1.03 – Prompt Engineering →