ORM

ORM stands for Object-Relational Mapping.

It’s a tool/library that lets you work with a database using programming-language objects instead of writing SQL for everything.

Without an ORM

Suppose you have a users table:

SELECT * FROM users WHERE id = 5;

You write SQL directly and get database rows back.

With an ORM

You might write something like:

const user = await prisma.user.findUnique({
  where: { id: 5 }
});

The ORM translates this into the appropriate SQL behind the scenes.

Your code
   ↓
ORM
   ↓
SQL
   ↓
PostgreSQL
   ↓
Rows
   ↓
ORM
   ↓
Your objects

Why use an ORM?

It makes common database operations easier:

Create user
Find user
Update user
Delete user
Find user's tasks
Create project

Instead of repeatedly writing SQL, you use the ORM’s API.

Examples of ORMs

For the technologies you’ve been working with:

• Prisma → Node.js/TypeScript

• Sequelize → Node.js

• TypeORM → TypeScript/JavaScript

• Hibernate → Java

For example, with Prisma:

const users = await prisma.user.findMany();

instead of:

SELECT * FROM users;

ORM and relationships

This is one of the useful parts.

Suppose:

User
 │
 └── has many
       ↓
     Tasks

An ORM can let you express that relationship directly:

const user = await prisma.user.findUnique({
  where: { id: 5 },
  include: { tasks: true }
});

Instead of manually writing the SQL joins.

But ORM ≠ database

This is important:

PostgreSQL = database
Prisma     = ORM

The ORM doesn’t replace PostgreSQL.

It sits between your application and the database:

Express application
       ↓
     Prisma
       ↓
   PostgreSQL

Mental model:

ORM = a translator between your application’s objects/code and relational database tables/SQL.

---

Database My-Journey-In-Codeless