Regal
Regal is an AI-powered personal stylist for Persian-speaking users. People photograph the clothes they already own, Regal turns them into a digital wardrobe, analyses their style and colors, and suggests complete outfits for the occasion and the day's weather. Radical designed and built the product end to end: the Flutter app, the Django backend and the computer-vision models.
At a glance
Regal is an AI stylist and digital wardrobe app in fashion and lifestyle, built by Radical in 2025 using Flutter, Django REST Framework, PyTorch, Transformers, TensorFlow Lite, Firebase.
| Product | AI stylist and digital wardrobe app |
|---|---|
| Industry | Fashion and lifestyle |
| Audience | Men and women in Iran who want to dress better with what they own |
| Platforms | Android and iOS app, website regalapp.ir |
| Language | Persian |
| Radical's role | Product design, mobile app, backend, AI models, deployment |
| Stack | Flutter, Django REST Framework, PyTorch, Transformers, TensorFlow Lite, Firebase |
| Year | 2025 |
The challenge
Most people stand in front of a full closet every morning and still feel they have nothing to wear. Style advice usually means buying more. The goal was a consumer app that makes expert styling feel effortless, works with the clothes people already own, and keeps personal photos private.
What we built
- Digital wardrobe from photos: users photograph each item and Regal catalogs it automatically.
- Wardrobe analysis: dominant styles and colors are identified using color theory, so users understand their own taste.
- Goal and occasion styling: users say what they need (a meeting, a party, a casual day) and get complete outfits.
- Weather-aware suggestions built from items already in the closet.
- Separate experiences for men and women, each with its own priorities.
- Smart shopping advice: brand recommendations based on what the wardrobe is missing, so users buy less and wear more.
- Credit system with free starter credits, so users see results before paying.
- Privacy first: photos and data are stored encrypted.
Technology
The mobile app is built in Flutter with Firebase for notifications and analytics, and runs lightweight models on the device with TensorFlow Lite. A Django REST backend handles accounts, wardrobes and credits. A separate AI service built on PyTorch and Transformers, with its own fine-tuning pipeline and GPU deployment, classifies clothing and generates outfit combinations.
Result
Regal shows how Radical turns a complex AI idea into a simple consumer product: from branding and UX to computer vision, a production backend and app-store releases, delivered by one team.
Planning a mobile app with AI inside? See our services or tell us about your idea for a free estimate.