The problem
The client wanted a food photo to be the starting point for a meal record, reducing manual entry.
← Selected work
Consumer AI
Recipe Lens
Recipe Lens turns a food photo into suggested ingredients, recipe information and estimated calories and macronutrients. QPOI built the iOS and Android app, from camera capture and image analysis to saved meal history and subscriptions.
- Sector
- Consumer AI
- Platform
- iOS + Android
- Stack
- React Native (Expo)
- Duration
- 3 months
- Engagement
- MVP engineering
- Status
- Launched
The project at a glance
What QPOI delivered
QPOI connected photo capture to suggested ingredients and nutrition estimates, saved history and subscriptions.
Delivery evidence
An iOS and Android MVP was delivered. Completed meal records and images remain available locally.
Photo-to-nutrition pipeline, as builtView diagramHide diagram
1.0
The challenge
The client wanted a food photo to be the starting point for a meal record. Instead of entering each detail manually, users would receive suggested ingredients, recipe information and nutrition estimates from the image.
The engineering work connected image capture to model analysis, then converted the response into fields the app could display and save. Images also needed to remain manageable on a phone, and saved meal records needed to survive interrupted connectivity.
2.0
Core features
2.1
Suggested ingredients and recipe information
A vision model returns likely ingredients, a meal description and recipe information. The app converts that response into a structured meal record.
2.2
Nutrition estimation from images
Photo analysis produces estimated calories and macronutrients alongside the meal description. These are model estimates from the image, rather than measured nutritional values.
2.3
Local-first storage and meal history
Captured images and completed analysis results are saved in local meal history. Users can revisit saved records without repeating the image analysis.
2.4
Optimised image handling
Photos are resized and compressed on the phone before upload, reducing the payload sent for analysis.
3.0
Mobile engineering fundamentals
3.1
Local data handling
Local file and record management keeps saved photos associated with their meal descriptions and nutrition estimates.
3.2
Subscription logic
Subscription handling tracks plan state and usage caps, including restoring purchases.
3.3
Permissions and compliance
Managed permissions for camera and photo library, with compliant IAP restore flows.
4.0
Outcome
- Delivered an MVP for iOS and Android
- Connected camera and photo-library input to structured meal analysis
- Integrated subscriptions, purchase restoration and usage limits
- Saved completed meal records and images locally for later review
Built with
- React Native (Expo)
- TypeScript
- Gemini 1.5 Flash (Vision)
- RevenueCat
- Expo FileSystem
- AsyncStorage
Next project
Build IG