Virtual Wardrobe is a full-stack, cloud-native application designed to help users efficiently manage their personal wardrobe, reduce overconsumption, and make sustainable fashion choices. Version 2.0 brings a modern, scalable architecture across both frontend and backend, ensuring a smoother, faster, and more immersive user experience.
| Component | Old Stack | New Stack |
|---|---|---|
| Frontend | HTML, CSS, JavaScript | Next.js (TypeScript) + Tailwind CSS |
| Backend Framework | Django (synchronous) | FastAPI (asynchronous, OpenAPI support) |
| ORM | Django ORM | Prisma ORM (type-safe, auto-generated client) |
| Database | SQLite | PostgreSQL via Neon Serverless PostgreSQL |
| Cloud Provider | AWS | Google Cloud Platform (VMs, Cloud Storage) |
| Authentication | Django Session-based | JWT (stateless) + Google OAuth 2.0 |
| Architecture | Monolithic | Microservices |
| Caching | Not implemented | Redis (in-memory caching) |
| Database Indexing | Not implemented | Indexed key tables for optimized query performance |
| Virtual Try-On | Not implemented | Virtual Try-on API (Vertex AI) |
| Logging & Monitoring | Not implemented | Grafana Loki + Dashboards for log aggregation |
The frontend was completely re-engineered in Version 2.0 using Next.js with TypeScript and Tailwind CSS, offering a modern, responsive, and performant UI. Key improvements:
- Server-side rendering for faster initial load times and SEO benefits.
- Tailwind CSS for utility-first, maintainable styling.
- Dynamic outfit filtering and search for quick discovery.
- Virtual Try-On integration with the backend.
- Mobile-first design approach for an optimized user experience across devices.
This revamp replaces the basic HTML, CSS, and JS used in the first version.
- Asynchronous, high-performance API framework.
- Auto-generates OpenAPI (Swagger) docs.
- Fast response times and better scalability.
- Type-safe and auto-generated client.
- Simple schema definitions and DB migrations.
- Clean data modeling for users, items, and outfit combinations.
- Replaces SQLite with serverless PostgreSQL for scalability.
- Auto-scaling capabilities to handle fluctuating traffic.
- Speeds up browsing and reduces backend load.
- Cached filters, outfits, and frequently queried data.
- Cloud VMs for compute.
- Cloud Storage for static/media assets.
- Enables CI/CD and scalable deployments.
- Stateless JWT-based authentication.
- Integrated Google OAuth 2.0 for SSO.
- Scalable and secure.
- Powered by the Virtual Try-on API provided by Vertex AI.
- Users can visualize outfits directly in the app.
- Real-time logs and system health monitoring.
- Debug and analyze across microservices.
- Microservices-based design for modularity and scalability.
- Stateless services with dedicated caching, logging, and storage layers.
- Cloud-optimized deployment with environment-based configurations.
- 40% faster API responses via async FastAPI and Redis caching.
- Improved data management with Prisma and PostgreSQL.
- Seamless authentication and smoother login flows.
- Faster page loads and enhanced frontend responsiveness.
- Better debugging and monitoring with centralized logging.
Virtual Wardrobe empowers users to:
- Track their fashion inventory.
- Make mindful decisions around purchases.
- Embrace a minimal, sustainable lifestyle through intelligent recommendations and outfit planning.
If you're interested in exploring or contributing to the project:
- 📬 Connect on LinkedIn
- 💻 View the full codebase on GitHub