Prisma ORM

Prisma gives TypeScript a type-safe database client generated from your schema. Model once, get autocomplete everywhere.

45 min · Reviewed 2026

Schema Is Source of Truth

You write one `schema.prisma` file. Prisma generates a TypeScript client whose types match exactly. Renaming a column is safe because everything that touches it fails to compile.

// schema.prisma
datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")
}

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

model User {
  id        String   @id @default(cuid())
  email     String   @unique
  createdAt DateTime @default(now())
  posts     Post[]
}

model Post {
  id       String  @id @default(cuid())
  title    String
  body     String
  author   User    @relation(fields: [authorId], references: [id])
  authorId String
}Run `npx prisma migrate dev --name init` to create tables. Run `npx prisma generate` to refresh the client.

import { PrismaClient } from "@prisma/client";
const prisma = new PrismaClient();

export async function createPost(authorEmail: string, title: string, body: string) {
  try {
    return await prisma.post.create({
      data: {
        title,
        body,
        author: { connect: { email: authorEmail } },
      },
      include: { author: true },
    });
  } catch (e) {
    console.error("createPost failed", e);
    throw e;
  }
}

export async function recentPosts() {
  return prisma.post.findMany({
    orderBy: { id: "desc" },
    take: 20,
    include: { author: { select: { email: true } } },
  });
}`include` brings relations. `select` picks fields. Types are exact for each query.

Understanding "Prisma ORM" in practice: AI-assisted coding shifts work from syntax recall to design thinking — models handle boilerplate so you focus on architecture. Prisma gives TypeScript a type-safe database client generated from your schema. Model once, get autocomplete everywhere — and knowing how to apply this gives you a concrete advantage.

Apply schema in your ai-coding workflow to get better results
Apply migration in your ai-coding workflow to get better results
Apply client in your ai-coding workflow to get better results
Apply relation in your ai-coding workflow to get better results

Use AI to generate unit tests for an existing function
Ask AI to refactor a messy function and explain the changes
Have AI suggest a code review for a recent pull request

The big idea: schema is truth, the client is type-safe, and a good ORM makes wrong queries fail at compile time.

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-progx-prisma-creators

In Prisma, what is considered the 'source of truth' for your database structure?
1. The database itself via direct SQL
2. The schema.prisma file
3. The generated TypeScript client
4. The migration history folder
What happens when you rename a field in your schema.prisma file?
1. Prisma ignores the change until you regenerate
2. All code using the old field name will fail to compile
3. You must manually update all query files
4. Prisma automatically renames the database column
What does Prisma generate from your schema.prisma file?
1. A TypeScript client with matching types
2. An SQL script that runs on startup
3. A Docker container
4. A GraphQL API endpoint
By default, what fields does a Prisma query return?
1. Nothing until you specify include
2. Only the fields specified in the select clause
3. Every scalar field defined in the model
4. Only the ID and foreign keys
In a performance-critical code path, what should you use to optimize a Prisma query?
1. Default query settings
2. The include option to fetch relations
3. Raw SQL queries
4. The select option to return only needed fields
Why is calling prisma inside a loop inefficient?
1. Loops are not allowed in Prisma queries
2. Each iteration creates a new database connection
3. Prisma prevents concurrent connections
4. It causes N+1 query problems with separate database round-trips
What should you use to batch multiple database operations in Prisma?
1. The batch option
2. Promise.all
3. A for loop
4. prisma.$transaction
What is the purpose of the 'include' option in a Prisma query?
1. To exclude certain fields from results
2. To return only specific scalar fields
3. To fetch related records in a single query
4. To run a raw SQL query
What type of errors occur when you use a field name that doesn't exist in your Prisma schema?
1. Compile-time TypeScript errors
2. Runtime SQL errors
3. Network timeout errors
4. Authentication errors
In Prisma, what is a migration?
1. A versioned change to your database schema
2. A way to move data between databases
3. A type of Prisma query
4. A backup of your data
How does Prisma provide type safety for database operations?
1. Through runtime type checking
2. Via database constraints
3. Using JSON validation
4. By generating TypeScript types from the schema
What happens if you try to select a field you didn't include in your select clause?
1. You get a compile-time error
2. The query returns all fields anyway
3. You get a runtime exception
4. The query returns undefined for that field
Why is compile-time type checking valuable for database queries?
1. Errors are caught before deployment, not in production
2. It makes queries run faster
3. It generates better SQL
4. It creates the database automatically
What is the benefit of the schema being the 'single source of truth'?
1. You don't need backups
2. TypeScript is not required
3. The database runs faster
4. Changes only need to be made in one place
What occurs at compile time with a well-typed ORM like Prisma when a query references a deleted database column?
1. The code runs normally with null values
2. TypeScript shows an error before runtime
3. The database returns an error
4. Nothing happens

← Back to interactive lesson

Tendril · Creators · AI-Assisted Coding

Prisma ORM

Prisma gives TypeScript a type-safe database client generated from your schema. Model once, get autocomplete everywhere.

45 min · Reviewed 2026

Schema Is Source of Truth

You write one `schema.prisma` file. Prisma generates a TypeScript client whose types match exactly. Renaming a column is safe because everything that touches it fails to compile.

// schema.prisma
datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")
}

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

model User {
  id        String   @id @default(cuid())
  email     String   @unique
  createdAt DateTime @default(now())
  posts     Post[]
}

model Post {
  id       String  @id @default(cuid())
  title    String
  body     String
  author   User    @relation(fields: [authorId], references: [id])
  authorId String
}Run `npx prisma migrate dev --name init` to create tables. Run `npx prisma generate` to refresh the client.

import { PrismaClient } from "@prisma/client";
const prisma = new PrismaClient();

export async function createPost(authorEmail: string, title: string, body: string) {
  try {
    return await prisma.post.create({
      data: {
        title,
        body,
        author: { connect: { email: authorEmail } },
      },
      include: { author: true },
    });
  } catch (e) {
    console.error("createPost failed", e);
    throw e;
  }
}

export async function recentPosts() {
  return prisma.post.findMany({
    orderBy: { id: "desc" },
    take: 20,
    include: { author: { select: { email: true } } },
  });
}`include` brings relations. `select` picks fields. Types are exact for each query.

Apply schema in your ai-coding workflow to get better results
Apply migration in your ai-coding workflow to get better results
Apply client in your ai-coding workflow to get better results
Apply relation in your ai-coding workflow to get better results

Use AI to generate unit tests for an existing function
Ask AI to refactor a messy function and explain the changes
Have AI suggest a code review for a recent pull request

The big idea: schema is truth, the client is type-safe, and a good ORM makes wrong queries fail at compile time.

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-progx-prisma-creators

In Prisma, what is considered the 'source of truth' for your database structure?
1. The database itself via direct SQL
2. The schema.prisma file
3. The generated TypeScript client
4. The migration history folder
What happens when you rename a field in your schema.prisma file?
1. Prisma ignores the change until you regenerate
2. All code using the old field name will fail to compile
3. You must manually update all query files
4. Prisma automatically renames the database column
What does Prisma generate from your schema.prisma file?
1. A TypeScript client with matching types
2. An SQL script that runs on startup
3. A Docker container
4. A GraphQL API endpoint
By default, what fields does a Prisma query return?
1. Nothing until you specify include
2. Only the fields specified in the select clause
3. Every scalar field defined in the model
4. Only the ID and foreign keys
In a performance-critical code path, what should you use to optimize a Prisma query?
1. Default query settings
2. The include option to fetch relations
3. Raw SQL queries
4. The select option to return only needed fields
Why is calling prisma inside a loop inefficient?
1. Loops are not allowed in Prisma queries
2. Each iteration creates a new database connection
3. Prisma prevents concurrent connections
4. It causes N+1 query problems with separate database round-trips
What should you use to batch multiple database operations in Prisma?
1. The batch option
2. Promise.all
3. A for loop
4. prisma.$transaction
What is the purpose of the 'include' option in a Prisma query?
1. To exclude certain fields from results
2. To return only specific scalar fields
3. To fetch related records in a single query
4. To run a raw SQL query
What type of errors occur when you use a field name that doesn't exist in your Prisma schema?
1. Compile-time TypeScript errors
2. Runtime SQL errors
3. Network timeout errors
4. Authentication errors
In Prisma, what is a migration?
1. A versioned change to your database schema
2. A way to move data between databases
3. A type of Prisma query
4. A backup of your data
How does Prisma provide type safety for database operations?
1. Through runtime type checking
2. Via database constraints
3. Using JSON validation
4. By generating TypeScript types from the schema
What happens if you try to select a field you didn't include in your select clause?
1. You get a compile-time error
2. The query returns all fields anyway
3. You get a runtime exception
4. The query returns undefined for that field
Why is compile-time type checking valuable for database queries?
1. Errors are caught before deployment, not in production
2. It makes queries run faster
3. It generates better SQL
4. It creates the database automatically
What is the benefit of the schema being the 'single source of truth'?
1. You don't need backups
2. TypeScript is not required
3. The database runs faster
4. Changes only need to be made in one place
What occurs at compile time with a well-typed ORM like Prisma when a query references a deleted database column?
1. The code runs normally with null values
2. TypeScript shows an error before runtime
3. The database returns an error
4. Nothing happens

← Back to interactive lesson