AI Spam Call Blocker App

Cliente Freelancer · Remoto · Remoto · freelance · mid · 1500–12.500 INR

Publicada el 2026-07-31

Descripción de la oferta

I need a clean-looking, lightweight mobile application that shields users from unwanted phone calls on both iOS and Android. The core of the solution is an AI-driven engine that learns to recognise spam numbers and patterns in real time rather than relying on static blacklists. Here is what I am aiming for: • Dual-platform delivery – a single code-base (Flutter, React Native or Kotlin + Swift if you prefer native) that compiles smoothly to the latest iOS and Android versions, passes App Store and Google Play review guidelines, and makes use of each platform’s allowed call-screening APIs. • AI identification – the app should run an on-device model or securely reach a cloud endpoint to analyse caller ID metadata and behaviour. I am open to TensorFlow Lite, Core ML, or a compact custom model as long as latency is minimal and privacy is preserved. • Automatic blocking – when the model flags a call as spam the user never hears a ring; the call is silently rejected and logged inside the app with date, time and reason. • Simple activity log – a scrollable list showing blocked numbers, confidence score and an option to mark “Not spam” so the model keeps improving. • Minimal permissions – stick to phone access only; no contact harvesting, GPS, or other extras that could cause distrust during review. I’ll supply brand colours and basic UI wireframes; you handle end-to-end development, model integration, unit tests, and packaging for TestFlight / internal track. Acceptance criteria: 1. 95 % of known spam numbers in a provided validation set are blocked. 2. Less than 1 % false positives in test calls from whitelisted numbers. 3. App passes submission review on both stores. If you have shipped call-screening or on-device ML features before, even better—tell me which tools you used and how you balanced accuracy with battery impact.

Skills

Fuente original: freelancer