Using Prisma with Postgres and Node.js

July 12, 2021

Most modern day applications make use of databases mainly to store data. In many modern backend architectures, we model everything else around the data and its manipulation.

When using Node.js most developers opt to go for NoSql databases as they are known to be easier to use. Examples of such are Mongodb and Firebase.

Working with SQL databases in Node.js often requires ORM (Object Relational Mapping) libraries to abstract the untidy database operations so the developer only has to worry about the logic.

In this article, we will look at how to setup and create a simple Node.js API, using Express as our framework of choice, Postgresql as our database, and Prisma as our ORM.


To comfortably follow this article along, you will need to have:

  • Node.js installed on your machine.
  • Postgresql installed on your machine.
  • A good understanding of the Express framework.
  • A working knowledge of postgress.

Step 1: Setting up the Application

Create your working directory and give it any name you’d like. I will call mine section-prisma. This is where all the magic will happen.

Run npm init -y in the directory to initialize a Node.js project.

 mkdir section-prisma

 cd section-prisma

 npm init -y

With that done, we can now install Prisma.

To do so, we use npm or yarn.

npm install prisma

The next step is to initialize prisma into the project.

In the terminal, we run:

npx prisma init

This command will create a folder within the root of your project called prisma.

Within this folder is a file called schema.prisma. This file will contain our configuration for the prisma in this project. The command also creates a .env file.

A .env file is typically used to store sensitive values you wouldn’t want public like keys and secret tokens or even encryption secrets. It is best practice to add it in your .gitignore file.

Step 2: Connecting to Postgres

This article already assumes that you have postgress already setup on your machine and we will not go through that.

Prisma supports a lot of databases, and postgress is just one of them. In the event that you want to use another database like sqlite, you can change the value of the database provider in the datasource option in the schema.prisma file.

Since we are using postgres, our schema.prisma file will look like this:

datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")

generator client {
  provider = "prisma-client-js"

Now in the .env file, we will add our Database URL.

For postgress, it usually looks like this:


Replace, your_user, your_password and db_name with your correct credentials.

With that done, we have our project setup.

Let’s get coding!

Step 3: Data modeling using Prisma

When working with a database, you have to define how the data in the database will look like. Modeling is necessary to give a description of the data that the client is providing, and the data the database is expecting.

Prisma comes with it’s own data modeling language that makes it easier to perform data modeling.

Prisma is an ORM, so it has to transform the data model into relevant SQL statements and database tables. Therefore, once we are done with our models, we will run prisma migrate which will create the relevant tables in the db.

Step 4: Creating our model

Within the schema.prisma file, after the datasource and generator block, we can define our model.

TIP: If you are using VSCode for this tutorial, you can install the prisma extention here to help you autocomplete and format the prisma modeling language.

To define a model, we use the keyword model then specify the name of the model. This entire model will be treated as a table by the database.

  model Job { }

We can now add fields into our model. Prisma models use the same db primitives and types used in normal models.

In this article, we will use just one model for demonstration.

Within the Job model let’s define our fields.

model Job {
  id       Int    @id @default(autoincrement())
  title    String
  location String
  salary   String

We defined title, location and salary fields which all are of the String type. We also have an id so that each Job can be unique.

In prisma, we can define attributes like @id which denotes that this field will be treated as an id, and @default() which sets the default value for the field.

Some attributes take arguments similar to the way default took the autoincrement() argument.

And just like that, you’ve built your first Prisma model.

We should now sync it with our db by running:

npx prisma migrate dev

The command will prompt you to enter a name for the migration, so you should be very intentional with the data you provide.

Once the command is done, you should see a migrations folder in the Prisma folder containing the migration we just made to the db.

You can read more on modeling data in prisma in the official docs

Step 5: Creating a Rest API

We can now query the db while creating an express API around our data.

Creating an Express Server

First step, let’s create our server using express.

To install express run:

 npm i express

In the root of the folder, create a file called index.js, that is where all the magic will happen.

In the file, we will create a simple server:

const express = require("express");

const app = express();


const port = process.env.PORT || "3000";

app.listen(port, () => {
  console.log(`Server Running at ${port} 🚀`);

One of the key features that Prisma ships with is querying and query relations.

It provides some query options that help you perfom CRUD operations and other operations. We will use a few in our example. To read more about them, you can check the docs

We can create a route to create jobs. It will be a post request that expects a body with a job object.

const { PrismaClient } = require('@prisma/client');
const prisma = new PrismaClient();

...'/job', async (req, res) => {
  const job = await prisma.job.create({ data: req.body });

In the code snippet above, we imported the PrismaClient, then created a variable to initialize the client.

Within the POST method, we used the create query from prisma to create a new record in the database, then passed in the request body to the data object.

We can now create a route to get all the jobs. It will be a GET request which finds all jobs and returns them as json.

const { PrismaClient } = require('@prisma/client');
const prisma = new PrismaClient();

app.get('/', async (req, res) => {
  const job = await prisma.job.findMany();

You can try to finish up the UPDATE and DELETE methods on your own to get familiar with the ORM.

The final code should looks something like this:

const { PrismaClient } = require("@prisma/client");
const express = require("express");
const prisma = new PrismaClient();

const app = express();

app.use(express.json());"/job", async (req, res) => {
  const job = await prisma.job.create({ data: req.body });

app.get("/", async (req, res) => {
  const job = await prisma.job.findMany();

const port = process.env.PORT || "3000";

app.listen(port, () => {
  console.log(`Server Running at ${port} 🚀`);

With that, you just created a simple, functioning API using Node.js (Express), Postgres and PRISMA. 🙌


Prisma is a very powerful ORM. It offers a lot of features out the box, but the key feature is the ease of use in comparison to other ORMs in place.

In this article we covered:

  • The basics of Prisma.
  • How to setup Prisma and connect to Postgres.
  • How to create an API to query the DB.

It’s also worth noting that commands like prisma migrate are more powerful than demonstrated in the article and are worth giving a good look into.

You can read more about that here

Prisma does not support Postgres alone. It supports multiple SQL databases so feel free to experiment with your favorite DB.

Happy coding!

Peer Review Contributions by: Linus Muema