Paymada analyzes payment SMS and push notifications on the user's phone in real time, delivering instant point rewards and gamified benefits (lottery / scratch cards) every time a payment occurs.
Client
Paymada
Year
2026
Scope
Android Development (solo)
Platform
Android · Kotlin · Coroutine · Flow · Compose · MVI
The Challenge
Detecting payment notifications reliably across 90+ Korean and international finance apps, managing real-time payment-data state without regressions, building a gamified reward / rewarded-ad pipeline, securing sensitive financial data, and compressing launch lead-time.
Our Solution
Built an AI-orchestrated dev pipeline with Claude Code (spec → MVI architecture → TDD code-gen → auto-review); implemented a SmsReceiver + NotificationListenerService payment-parsing engine with a regex-based, general-purpose PaymentParser; adopted a strict Contract-based MVI unidirectional architecture with Jetpack Compose for declarative UI; wired AdMob rewarded ads into a scratch-card game controlled by Firebase Remote Config; and hardened security with EncryptedSharedPreferences + an AuthInterceptor for automatic JWT refresh.
Results
Cut end-to-end launch lead-time by 30–40% through AI orchestration, covered 90+ domestic and international finance apps in the parsing engine, sharply reduced state-management bugs via MVI, improved reward participation for newly acquired users, and slightly lowered pre-release bug rate through AI-driven TDD.