How Our AI Works
At FitWardrobe, we believe you should know exactly what happens when you click "Generate Outfit" or "Analyze Photo."
The Technology Stack
We use a hybrid AI architecture to balance speed, intelligence, and privacy.
Google Gemini 2.0 Flash
Powers our "Snap & Tag" feature. It looks at your photos to identify clothing items with high precision.
- Input: Garment Photographs
- Output: Tags (Type, Color, Style)
- Privacy: Temporary processing, no storage.
Groq AI (Llama 3 Powered)
Powers our styling engine and chat assistant. It takes your digital tags and reasons like a personal stylist.
- Input: Wardrobe Tags + Chat Context
- Output: Recommendations & Advice
- Privacy: No data used for training.
⚠️ What AI Can & Cannot Do
✅ What it's great at:
- Organizing 100+ items in seconds
- Suggesting fresh color combinations
- Reminding you of clothes you forgot you had
- Building capsule wardrobes efficiently
❌ Where it can fail:
- Identifying hyper-local Indian ethnic wear
- Understanding fabric "feel" through photos
- Accounting for real-time weather changes
- Perfectly matching your body type nuances
Common Questions
Is the AI training on my photos?
No. We specifically use API configurations that prevent our providers (Google and Groq) from using your data to train their models. Your wardrobe is your business.
Can FitWardrobe employees see my clothes?
Unless you explicitly send us a support screenshot, no. Your wardrobe data is stored in a local database on your own device. We only store your email address for account sync.
Why use two different AI systems?
Gemini is a "multimodal" expert - it's the best at seeing. Groq is a "reasoning" expert - it's the best at thinking fast. By combining them, we give you a stylist that's both smart and lightning fast.
Help Us Improve
Our AI is in Beta. If it makes a mistake, tell us!
Send Feedback