Activity Monitoring AI Assistant
The Activity Monitoring AI Assistant is a Flutter-based mobile application that tracks app usage statistics, stores the data in a local SQLite database, and provides AI-powered insights. The app runs a background service to log app usage and allows users to ask questions about their activity using an AI assistant.
Activity Monitoring AI Assistant
AI-Powered Mobile Activity Tracker
The Activity Monitoring AI Assistant is a Flutter-based mobile application that tracks app usage statistics, stores data locally in SQLite, and provides AI-powered insights about digital habits. The app runs with minimal permissions and keeps all data on-device.
Problem
People want to understand their digital habits but most tracking apps either require invasive permissions, send data to cloud servers, or provide only basic statistics without meaningful insights. There is a gap between simple screen-time counters and genuinely useful behavioral analysis.
Solution
This app tracks app usage at the system level, stores all data locally in SQLite, and uses on-device AI to generate insights about usage patterns. No data leaves the device. The AI layer identifies trends like increasing screen time, most-used apps at different times of day, and usage patterns correlated with specific activities.
My Role
I built the entire application: the Flutter frontend, the native platform channels for usage data collection, the SQLite data layer, and the AI insight engine. I also designed the data model for efficient storage and querying of usage statistics over time.
Technical Architecture
The app uses Flutter for the UI layer with platform-specific native code for accessing usage statistics on Android. Data is stored in a local SQLite database with a schema optimized for time-series queries. The AI layer processes usage data locally to generate insights without requiring network access or cloud processing.
Key Features
Usage Tracking
Tracks app usage duration, frequency, and session patterns. The tracking runs as a background service with minimal battery impact.
Local Data Storage
All data is stored in SQLite on the device. No cloud sync, no network requests for user data. The database schema supports efficient aggregation queries for daily, weekly, and monthly reports.
AI-Powered Insights
On-device AI analyzes usage patterns to identify trends, anomalies, and correlations. Insights are presented as natural language summaries and visual charts within the app.
Privacy by Design
The app requires minimal permissions and transmits zero user data externally. All processing happens on-device, making it suitable for users who prioritize privacy.
Tech Stack
Flutter, Dart, SQLite, LLM, Native Platform Channels
Outcome
The project delivered a functional privacy-first activity tracker with AI-powered insights. It demonstrates my ability to work with native platform APIs, design efficient local data architectures, and integrate AI capabilities into mobile applications while maintaining strict privacy boundaries.