Introduction to AI APIs: Add AI to Your Own Apps

So far you have used AI through chat apps. To build your own AI-powered tools — a support assistant on your website, an app that writes product descriptions — you use an API.

What is an API?

An API (Application Programming Interface) is a way for one program to talk to another. Think of a restaurant: you (the app) give your order to the waiter (the API), the kitchen (the AI model) prepares it, and the waiter brings back your food (the response).

How an AI API call works

Your server sends a request with your prompt and settings.
The request includes an API key that identifies your account.
The AI provider runs the model and returns a response, usually in JSON format.
Your app shows or uses the result.

The golden rule: protect your API key

An API key is like a password linked to your bill. Never put it in website JavaScript, mobile app code or a public GitHub repository. Keep it on your server in an environment variable, and set spending limits in your provider dashboard.

Costs

AI APIs are usually charged by usage (often per token). Start small, monitor usage, and cache results that don’t need to be regenerated.

What you can build next

A FAQ assistant that answers from your own documents
A tool that turns product photos into descriptions
An automation that summarises daily reports

Every major AI provider publishes official documentation with exact request formats — always follow the latest docs for the provider you choose.

A generic AI API call (check your provider’s docs for the exact format)

javascript

// Node.js example (server-side only — never put API keys in the browser)
// Install the provider's SDK, then set the key as an environment variable.

const apiKey = process.env.AI_API_KEY;

async function askAI(question) {
  const res = await fetch(process.env.AI_API_URL, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "Authorization": `Bearer ${apiKey}`
    },
    body: JSON.stringify({
      model: process.env.AI_MODEL,
      messages: [{ role: "user", content: question }]
    })
  });
  if (!res.ok) throw new Error("AI request failed: " + res.status);
  return res.json();
}

askAI("Suggest 3 names for a tea shop in Vadodara").then(console.log);