AI · Fashion Tech
Style Sense
From an empty closet to a daily outfit suggestion — powered by computer vision, weather, and personal style history.
Industry
Fashion Tech · Consumer AI
Year
2025
Duration
16 weeks
Team
1 product designer, 2 senior full-stack engineers, 1 ML engineer
Results
Outcomes, measured.
8×
Faster outfit decisions
From 4 minutes browsing to 30 seconds.
40k+
Garments catalogued
In the first 12 weeks post-launch.
92%
Tag accuracy
Auto-categorisation correct on first try.
4.7★
App Store rating
From 600+ reviews in launch quarter.
The challenge
What was broken.
- Users had to manually tag every clothing item — most gave up after the first 10. The catalogue was the entire product, but onboarding was killing it.
- Outfit suggestions felt random. The first version used a generic LLM with no understanding of weather, occasion, or what the user actually wore most.
- iOS and Android needed feature parity, but the team wanted to ship quickly without doubling engineering cost.
Our approach
How we solved it.
- Replaced manual tagging with computer-vision auto-categorisation. Drop a photo, get type / colour / style / formality back in 3 seconds.
- Built a hybrid recommendation engine: rules (weather, calendar, dress codes) gate an LLM that reasons over the user's style history and recent wear.
- Shipped a single React Native codebase for iOS + Android with native modules for camera capture, background image processing, and HealthKit integration.
- Set up production observability from week one — every recommendation logged with the user's accept / reject signal, feeding a continuous eval pipeline.
What we built
Concrete deliverables.
iOS + Android app (React Native + native camera modules)
Vision LLM pipeline for garment recognition and style scoring
Recommendation engine combining rules, retrieval, and an LLM agent
Outfit history with weekly insights and 'capsule wardrobe' analysis
Subscription billing via Stripe with weekly + annual plans
Admin tools for the in-house stylist team to curate trending looks
Tech stack
Next.jsReact NativeTypeScriptVision LLM (OpenAI)SupabasePostgreSQLStripeAWS S3Sentry
Services used
The capabilities behind this build.
AI Solutions
LLM agents, retrieval pipelines, and ML integrations that unlock real business leverage — not demos.
Read more
Mobile App Development
Native-grade iOS & Android apps with Flutter and React Native — crafted for scale, stability, and delight.
Read more
SaaS Product Development
Zero to revenue. Multi-tenant architecture, billing, auth, dashboards, analytics — done properly.
Read more
Want to ship something like this?
Tell us your problem. We'll come back with a plan, timeline, and fixed pricing.