Building an AI SaaS with Next.js, OpenAI, and Stripe in 2025

What We Are Building

A production-ready AI SaaS with Next.js App Router, OpenAI API, Stripe billing, and NeonDB PostgreSQL. This is the exact stack used to build Penline, a live AI PDF summariser.

Tech Stack

Next.js 14 App Router and Server Actions for the frontend and API layer
OpenAI API with GPT-4o for AI processing
Stripe for subscription billing and payment management
NeonDB for serverless PostgreSQL database
Tailwind CSS for styling

Step 1: Project Structure

Organise your project with clear separation between the app router pages, API routes for AI and webhooks, and lib files for your OpenAI, Stripe, and database clients.

Step 2: OpenAI Integration

Create a server-side API route that accepts text and calls the OpenAI chat completions endpoint. Never call OpenAI directly from client-side code as your API key will be exposed in the browser.

Step 3: Stripe Subscription

Use Stripe Checkout for the payment flow. Create a checkout session server-side, redirect the user to Stripe, then handle the webhook event to activate their subscription in your database after successful payment.

Step 4: Database with NeonDB

NeonDB gives you a serverless PostgreSQL database that scales to zero. Use it to store user accounts, subscription status, and usage data. Connect with the Postgres.js or Drizzle ORM library.

Step 5: Deploy to Vercel

Push to GitHub, connect to Vercel, add your environment variables for OpenAI, Stripe, and NeonDB. Your AI SaaS is live. Set up Stripe webhooks pointing to your production URL.

Building an AI product? Schedule a meeting and let us discuss.