Backend.md

Backend Development Guide

This guide documents the setup, execution, and structure of the PrepVerse Node.js/Express.js backend engine.

---

⚡ Server Entry Point (server.ts)

The backend resides in server.ts at the root directory. It runs as a self-contained Express server that serves two primary purposes:

1. API Proxying: Exposing /api/* endpoints to handle sensitive requests (like calling the Google Gen AI API) securely without exposing private environment variables to the browser.
2. Static Asset Delivery: In production (NODE_ENV === 'production'), it serves compiled frontend bundles from the dist/ directory.

---

🧭 Request Pipeline & Middlewares

1. Development Mode

In development mode, we mount the standard Vite Development Server as a middleware directly within Express:

if (process.env.NODE_ENV !== "production") {
  const vite = await createViteServer({
    server: { middlewareMode: true },
    appType: "spa",
  });
  app.use(vite.middlewares);
}

This enables features like hot reloading and TypeScript compilation in development while running everything on a single port (3000).

2. Production Mode

In production mode, Vite is bypassed. Express serves the statically built frontend from the /dist directory and routes all unresolved routes to /dist/index.html to allow client-side routing to work seamlessly:

const distPath = path.join(process.cwd(), 'dist');
app.use(express.static(distPath));
app.get('*', (req, res) => {
  res.sendFile(path.join(distPath, 'index.html'));
});

---

🔐 Environment variables

The server relies on the following variables:

• GEMINI_API_KEY: Required to authenticate with the Google Gen AI API for mock interviews and resume feedback loops.
• PORT: (Default: 3000) Configured via the hosting container environment. Do not override this port.
• NODE_ENV: Set to production in production environments.