Full-Stack SaaS
Magic Makers
Built a Consumer AI Image Platform from Scratch
- Role
- Full-stack developer
- Timeline
- Build to production
- Team
- Solo
- Live link
- join.magicmakers.ai ↗
- End-to-end
- frontend + backend
- Async
- generation pipeline
The problem
Magic Makers set out to turn ordinary family photos into themed illustrated characters through AI, aimed at parents rather than technical users — requiring a generation pipeline behind an interface simple enough for a non-technical audience, a large theme library, reliable image handling, and secure subscription billing.
My approach
Backend & data
Nest.js and Express API over PostgreSQL on Supabase, modeling users, uploads, themes, generation jobs, and outputs. Handled asynchronous generation so long-running jobs didn't block the interface.
Media handling
Secure upload, storage, and delivery for user photos and generated outputs via Firebase, with export paths for printing, cards, social sharing, and profile images.
Billing
Stripe subscription model covering plan management, entitlements, and feature gating.
Frontend
Figma-to-React/Next.js with Tailwind CSS built for a consumer audience — upload flow, theme browsing across a large library, generation states, gallery management.
System shape
Frontend
React, Next.js, TypeScript, Tailwind CSS
Backend
Nest.js, Express
Data
PostgreSQL, Supabase, Firebase
Infra
Vercel, Railway
AI
AI Integration
Stack
Frontend
- React
- Next.js
- TypeScript
- Tailwind CSS
Backend
- Nest.js
- Express
Data
- PostgreSQL
- Supabase
- Firebase
Infra
- Vercel
- Railway
AI
- AI Integration
Result
A complete consumer AI product delivered end to end — generation pipeline, media handling, subscription billing, and a non-technical user interface — built and deployed to production.