Image Processing Service

A backend-focused image processing service designed to demonstrate real-world backend engineering, cloud infrastructure, and performance optimisation.

This project prioritises scalability, cost-awareness, and clean system boundaries over visual complexity.

---

🧠 What This Service Does

• Authenticated users can upload images
• Images can be transformed (resize, rotate, grayscale, etc.)
• Image metadata is stored in a database
• Image files are stored privately in S3
• Images are delivered directly via a CDN (CloudFront)
• The backend is never in the hot path for image delivery

---

🏗️ High-Level Architecture

Client (React) → API (Express + TypeScript) → PostgreSQL (metadata) → Redis (rate limiting & caching) → AWS S3 (private image storage) → CloudFront CDN (global image delivery)

• The backend never serves image bytes
• Images are fetched directly from CloudFront
• S3 remains private using Origin Access Control (OAC)

---

✨ Features

Core

• Image upload with validation
• Image transformations using Sharp
• Metadata storage using PostgreSQL + Prisma
• JWT-based authentication
• Per-user image isolation

Image Delivery

• Private S3 bucket
• CloudFront CDN with OAC
• CDN-cached image delivery (no signed S3 URLs)
• Cold cache vs warm cache behavior handled by CDN

Performance & Safety

• Redis-based rate limiting
• Redis caching for image metadata
• Backend removed from the hot path for image delivery

---

🛠️ Tech Stack

Backend

• Node.js + Express
• TypeScript
• Prisma + PostgreSQL
• Redis
• Sharp
• JWT Authentication

Infrastructure

• AWS S3 (private bucket)
• AWS CloudFront (CDN with OAC)

Frontend

• React + TypeScript
• TanStack Query
• Minimal UI by design (backend-first project)

---

🖼️ Image Processing

Image transformations are handled using Sharp

Supported transformations:

• Resize
  • width: Target width in pixels
  • height: Target height in pixels
• Rotate
  • rotate: Rotation angle in degrees
• Format Conversion
  • format: Output image format (jpeg, png, webp)
• Compression
  • compress: Compression level to reduce file size
• Grayscale
  • grayscale: Convert image to grayscale
• Flip
  • flip: Vertical flip
• Mirror
  • mirror: Horizontal flip
• Blur
  • blur: Blur intensity
• Watermark
  • watermarkText: Text watermark applied to the image

Processed images are stored in S3 and delivered via CloudFront, allowing:

• Low latency delivery
• Reduced backend load
• Automatic edge caching

---

🔐 Security Decisions

• JWT-based authentication for API access
• Images are not publicly accessible via S3
• CloudFront uses Origin Access Control (OAC) to securely access S3
• AWS WAF intentionally not enabled (resume project, low attack surface)

This setup balances security, cost, and simplicity.

---

⚡ Performance Notes

• First request to an image is a cold cache miss and fetches from S3
• Subsequent requests are served from CloudFront edge locations
• Typical latency improvement observed: ~2s → <300ms

---

📌 Why This Project Exists

This project was built to go beyond CRUD apps and explore:

• Real backend performance constraints
• CDN-based architectures
• Practical cloud infrastructure decisions
• Production-style system design