JSON

What Is JSON?

JSON stands for JavaScript Object Notation.

Despite its name, JSON is not limited to JavaScript. It is widely used by Python, PHP, Java, JavaScript, mobile applications, AI applications, and many other technologies.

JSON is mainly used to represent and exchange structured data.

{
    "name": "John",
    "age": 25,
    "is_student": true
}

This represents information about a person.

Why Do We Need JSON?

Imagine a Python application communicating with a different application written in JavaScript.

Python and JavaScript have different internal data structures. They need a common format that both applications understand.

Python Application
        ↓
       JSON
        ↓
Other Application
        ↓
       JSON
        ↓
Python Application

JSON acts as a common language for exchanging structured information.

JSON Object

A JSON object is surrounded by curly braces: { }.

{
    "name": "John",
    "age": 25,
    "country": "India"
}

JSON stores information using key-value pairs.

"name": "John"
   ↑       ↑
  key    value

Here:

• name is the key.
• John is the value.

JSON Data Types

JSON supports several basic types of values.

{
    "name": "John",
    "age": 25,
    "price": 99.50,
    "active": true,
    "address": null
}

The main JSON types are:

• String
• Number
• Boolean
• Object
• Array
• null

JSON uses lowercase true, false, and null.

JSON Arrays

A JSON array stores multiple values and uses square brackets: [ ].

{
    "name": "John",
    "skills": [
        "Python",
        "NumPy",
        "Pandas"
    ]
}

The skills value contains multiple items.

JSON arrays are similar to Python lists.

JSON Array
    ↓
[ "Python", "NumPy", "Pandas" ]

Python List
    ↓
["Python", "NumPy", "Pandas"]

JSON vs Python

JSON looks very similar to Python dictionaries, but they are not exactly the same thing.

{
    "name": "John",
    "age": 25,
    "active": true
}

{
    "name": "John",
    "age": 25,
    "active": True
}

Notice the difference:

JSON:
true

Python:
True

JSON is a data format. A Python dictionary is a Python data structure.

JSON as a String

JSON can be represented as text.

json_data = '{"name": "John", "age": 25}'

Here, json_data is a Python string containing JSON.

Python does not automatically treat that string as a dictionary.

print(type(json_data))

<class 'str'>

We need to convert it into a Python object.

Convert JSON String to Python

Python provides the built-in json module for working with JSON.

import json

json_data = '{"name": "John", "age": 25}'

person = json.loads(json_data)

print(person)

json.loads() means: load JSON from a string.

JSON String
    ↓
json.loads()
    ↓
Python Dictionary

Now you can access the values normally:

print(person["name"])
print(person["age"])

John
25

Convert Python to JSON

The opposite operation is also common.

We can convert a Python dictionary into a JSON string using json.dumps().

import json

person = {
    "name": "John",
    "age": 25
}

json_data = json.dumps(person)

print(json_data)

{"name": "John", "age": 25}

Python Dictionary
       ↓
json.dumps()
       ↓
JSON String

loads() vs dumps()

These two functions are easy to confuse, so remember the direction.

person = json.loads(json_data)

json_data = json.dumps(person)

JSON
 ↓
loads()
 ↓
Python


Python
 ↓
dumps()
 ↓
JSON

Working With JSON Files

JSON is also commonly stored in files.

Example file:

user.json

Contents:

{
    "name": "John",
    "age": 25,
    "skills": [
        "Python",
        "AI"
    ]
}

Python can read this file using json.load().

import json

with open("user.json", "r") as file:

    user = json.load(file)

print(user["name"])
print(user["skills"])

Notice that this is load(), not loads().

json.load()
    ↓
Reads JSON from a file

json.loads()
    ↓
Reads JSON from a string

Writing JSON to a File

Python can also save a dictionary as a JSON file.

import json

user = {
    "name": "John",
    "age": 25,
    "skills": [
        "Python",
        "AI"
    ]
}

with open("user.json", "w") as file:

    json.dump(user, file, indent=4)

json.dump() writes Python data directly into a JSON file.

The indent=4 makes the file easier for humans to read.

JSON With Requests

This is where JSON becomes important for API development.

When you call an API with Requests, the server may return JSON.

import requests

response = requests.get(
    "https://httpbin.org/json"
)

data = response.json()

print(data)

Requests provides response.json() to convert the JSON response into Python data.

API
 ↓
JSON Response
 ↓
response.json()
 ↓
Python Dictionary / List

You normally do not need to call json.loads() yourself when Requests has already parsed the response for you.

Sending JSON With Requests

APIs also commonly expect JSON in the request body.

import requests

data = {
    "name": "John",
    "age": 25
}

response = requests.post(
    "https://httpbin.org/post",
    json=data
)

print(response.json())

The important part is:

json=data

Requests converts the Python data into JSON and sends it as part of the HTTP request.

Python Dictionary
       ↓
Requests
       ↓
JSON
       ↓
POST Request
       ↓
API

Nested JSON

Real APIs often return more complicated JSON structures.

{
    "user": {
        "name": "John",
        "age": 25,
        "address": {
            "city": "Hyderabad",
            "country": "India"
        }
    }
}

In Python, you can access nested values using multiple dictionary keys.

data["user"]["name"]

data["user"]["address"]["city"]

Example:

print(data["user"]["name"])
print(data["user"]["address"]["city"])

John
Hyderabad

JSON Arrays of Objects

APIs frequently return a list of objects.

{
    "products": [
        {
            "id": 1,
            "name": "Laptop",
            "price": 800
        },
        {
            "id": 2,
            "name": "Phone",
            "price": 500
        }
    ]
}

In Python:

products = data["products"]

for product in products:

    print(product["name"])
    print(product["price"])

Laptop
800
Phone
500

This pattern is extremely common when consuming APIs.

Why JSON Is Important for AI

Modern AI applications communicate with models and services through APIs. Those APIs commonly use JSON for requests and responses.

Python AI Application
        ↓
JSON Request
        ↓
AI API
        ↓
AI Model
        ↓
JSON Response
        ↓
Python Application

A simplified AI request might look like this:

{
    "model": "example-model",
    "messages": [
        {
            "role": "user",
            "content": "Explain Python"
        }
    ]
}

The exact structure depends on the API provider, but the important concept remains the same: structured data is exchanged as JSON.

Complete Python Example

Let's combine JSON conversion and API communication.

import requests
import json


data = {
    "name": "John",
    "age": 25,
    "skills": [
        "Python",
        "AI"
    ]
}


json_string = json.dumps(data)

print("JSON:")
print(json_string)


response = requests.post(
    "https://httpbin.org/post",
    json=data,
    timeout=10
)


response.raise_for_status()


result = response.json()

print("\nResponse:")
print(result)

The complete flow is:

Python Dictionary
       ↓
json.dumps()
       ↓
JSON String

Python Dictionary
       ↓
Requests
       ↓
POST Request
       ↓
API
       ↓
JSON Response
       ↓
response.json()
       ↓
Python Dictionary

Common Mistake

One common beginner mistake is confusing JSON with a Python dictionary.

user = {
    "name": "John",
    "active": True
}

{
    "name": "John",
    "active": true
}

They look almost identical, but Python and JSON have different rules.

The easiest way to remember this is:

Python Dictionary
       ≠
JSON

Python Dictionary
       ↕
Conversion
       ↕
JSON

What You Learned

• What JSON means
• Why applications use JSON
• JSON objects
• JSON arrays
• JSON data types
• JSON vs Python dictionaries
• json.loads()
• json.dumps()
• json.load()
• json.dump()
• Reading JSON files
• Writing JSON files
• Working with JSON through Requests
• Nested JSON
• Arrays of JSON objects
• Why JSON is important for AI APIs