Case study
NutriSnap
An Android food-recognition and nutrition diary application that classifies food from a photo using an on-device TensorFlow model and lets users track nutritional intake locally.
Overview
NutriSnap is an Android application for classifying food from images and tracking nutrition locally.
Problem
Manual nutrition logging can be tedious, especially when users need to enter food and serving information repeatedly.
Users
People who want a lightweight mobile diary for food recognition, serving-size entry and nutrition history.
Outcome
A mobile AI project demonstrating Android development, local persistence, image classification and nutrition-tracking UX.
Main capabilities
- Camera and gallery image selection.
- On-device TensorFlow food classification.
- Local diary history, serving-size entry and calorie charting.
My contribution
- Built Android application functionality around food recognition and diary workflows.
- Connected classification results to nutritional-value matching and local storage.
- Implemented local-history and progress-visualisation behaviours.
Technical approach
- Java and Android SDK provide the mobile application foundation.
- TensorFlow Android supports on-device food-image classification.
- SQLite stores diary entries locally.
- MPAndroidChart displays calorie progress in the app.
Important engineering decisions
- Used on-device classification to keep the core recognition flow local.
- Used SQLite for simple local persistence without requiring a remote service.
Challenges and trade-offs
- Mapping image classification output to useful nutrition entries.
- Keeping the diary workflow simple enough for repeated use.
Contact
I’m open to graduate and junior software, data and technology opportunities from late 2026.
Send me an email or connect with me on LinkedIn. I’m especially interested in Sydney and Melbourne roles across software engineering, fintech, full-stack product work, data and technology consulting.