HTML to PDF Serverless Service

A high-performance, serverless HTML-to-PDF conversion service built on AWS Lambda, designed to handle large-scale document generation for our fintech platform.

Problem

Our platform needed to generate thousands of PDFs daily — trade confirmations, account statements, settlement reports. The previous solution running on EC2 was:

• Expensive (always-on infrastructure)
• Hard to scale during peak hours
• Prone to memory leaks from Puppeteer instances

Solution

Migrated to a serverless architecture using AWS Lambda with Puppeteer Core:

Architecture

• AWS Lambda with custom Chromium layer for PDF rendering
• S3 for storing generated PDFs with pre-signed URLs
• API Gateway for REST endpoint access
• SQS for handling batch generation requests

Key Optimizations

• Cold-start optimization: Pre-loaded Chromium binary with custom Lambda layer
• Memory management: Careful browser instance lifecycle handling
• Concurrent processing: Stateless design supporting parallel PDF generation
• Template caching: Frequently used templates cached in Lambda's /tmp directory

Results

• 30% cost reduction compared to EC2 infrastructure
• Sub-3s response time for single PDF generation
• Auto-scales from 0 to thousands of concurrent executions
• Zero maintenance — no servers to manage

Technical Highlights

// Lambda handler with optimized Chromium launch
const chromium = require('@sparticuz/chromium');
const puppeteer = require('puppeteer-core');

exports.handler = async (event) => {
  const browser = await puppeteer.launch({
    args: chromium.args,
    executablePath: await chromium.executablePath(),
    headless: chromium.headless,
  });
  
  const page = await browser.newPage();
  await page.setContent(event.html, { waitUntil: 'networkidle0' });
  
  const pdf = await page.pdf({
    format: 'A4',
    printBackground: true,
  });
  
  await browser.close();
  return pdf;
};