i Feel Safe

i Feel Safe is a mission-critical personal security tool designed for immediate emergency response. Featuring custom low-level location filters, the app maintains an active foreground service on Android and continuous location streams on iOS.

RoleLead Security Architecture Engineer

PlatformsiOS, Android

Timeline8 Months (2022)

ArchitecturePlatform Channels

IndefiniteOS Survival Time

< 2.0sSOS Dispatch

2.8% / HrBattery Overhead

Tech Stack & Concepts

Platform ChannelsNative Background ServicesRedundant CommsGeofencing

Explore Repository

The Problem

Standard safety tracking applications are regularly put to sleep by aggressive OS battery-saving rules exactly when users need them most.

Product Thinking

Reliability in critical emergency moments is the absolute priority. The user must feel completely secure, knowing that an SOS dispatch will run even in deep sleep, low power, or zero network states.

Engineering Challenges

Bypassing native OS thermal and battery management profiles to run continuous background GPS tracking without triggering battery warnings.

Key Learnings

Mastered native iOS Swift background sessions, Android Kotlin foreground services, wake locks, and reliable location filtering algorithms.

Behind the Code

Architectural & Code Decisions

Deep technical breakdowns of individual structural choices and implementation patterns.

DECISION 01

Android Foreground Service & WakeLocks

Built an Android Foreground Service using Kotlin that runs with a persistent system notification, claiming CPU wake locks during active SOS states to completely prevent CPU throttling.

kotlin

// Safe Foreground location service setup
class EmergencyLocationService : Service() {
    private lateinit var wakeLock: PowerManager.WakeLock

    override fun onStartCommand(intent: Intent?, flags: Int, startId: Int): Int {
        val powerManager = getSystemService(Context.POWER_SERVICE) as PowerManager
        wakeLock = powerManager.newWakeLock(PowerManager.PARTIAL_WAKE_LOCK, "SafeApp::WakeLock")
        wakeLock.acquire(10*60*1000L /*10 mins*/)
        
        startForeground(NOTIFICATION_ID, createNotification())
        return START_STICKY
    }
}

DECISION 02

Fused Location Custom Filter

Implemented a Kalman-filter location algorithm inside the native layer to merge GPS, Cell-Tower, and Wi-Fi networks, ensuring high location accuracy.

swift

// High precision iOS location streamer
import CoreLocation

class SOSLocationStreamer: NSObject, CLLocationManagerDelegate {
    private let manager = CLLocationManager()
    
    func startSafetyStream() {
        manager.delegate = self
        manager.desiredAccuracy = kCLLocationAccuracyBestForNavigation
        manager.allowsBackgroundLocationUpdates = true
        manager.pausesLocationUpdatesAutomatically = false
        manager.startUpdatingLocation()
    }
}