Chapter 1: APIs and Integration

Connecting to AI Services

APIs are the gateway to integrating AI capabilities into your applications. Understanding how to work with AI APIs is essential for building AI-powered products.

---

Major AI APIs

Provider	Main API	Key Features
OpenAI	GPT-4, DALL-E	Most popular, comprehensive
Anthropic	Claude	Safety-focused, long context
Google	Gemini	Google integration
Hugging Face	Multiple	Open source models
Cohere	Command	Enterprise-focused

---

OpenAI API Basics

Authentication

from openai import OpenAI

client = OpenAI(api_key="your-api-key")

Basic Chat Completion

response = client.chat.completions.create(
    model="gpt-4",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "What is machine learning?"}
    ]
)

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

Key Parameters

Parameter	Purpose
model	Which model to use
messages	Conversation history
temperature	Creativity (0-2)
max_tokens	Response length limit

---

Anthropic (Claude) API

Basic Usage

import anthropic

client = anthropic.Anthropic(api_key="your-api-key")

message = client.messages.create(
    model="claude-3-sonnet-20240229",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Hello, Claude"}
    ]
)

---

Rate Limits and Quotas

Understanding Limits

• Requests per minute (RPM): API calls allowed
• Tokens per minute (TPM): Text volume allowed
• Tokens per day (TPD): Daily limits

Handling Rate Limits

import time
from openai import RateLimitError

def call_with_retry(func, max_retries=3):
    for attempt in range(max_retries):
        try:
            return func()
        except RateLimitError:
            time.sleep(2 ** attempt)
    raise Exception("Max retries exceeded")

---

Pricing Considerations

Token-Based Pricing

Model	Input	Output
GPT-4	$30/1M tokens	$60/1M tokens
GPT-3.5	$0.50/1M	$1.50/1M
Claude 3 Sonnet	$3/1M	$15/1M

Cost Estimation

Tokens per request × Requests per day × Days × Price per token

---

Security Best Practices

API Key Management

• Never commit keys to code
• Use environment variables
• Rotate keys regularly
• Use different keys per environment

import os
api_key = os.environ.get("OPENAI_API_KEY")

---

Key Takeaways

✅ APIs enable integration of AI into applications

✅ Different providers have different strengths

✅ Rate limits and costs require planning

✅ Secure key management is essential

---

Next Steps

Continue to Chapter 2: Building AI-Powered Apps.