Wizdom
Book, podcast & course summaries: 100K+ installs, 4.0+ rated
Wizdom delivers 15-minute book, course and podcast summaries in audio and text. Over two years I built the Flutter app that grew past 100K Play Store installs (4.0+ rating, also on iOS), the Firebase backend architecture behind it, and a Flutter Web admin panel that let a non-technical content team run the catalog.
The Problem
A content-heavy consumer app must feel instant: audio that plays offline, summaries synced across devices in real time, and content operations that don't need an engineer in the loop.
The Solution
A Flutter app with offline audio playback, dark mode, and content sharing; Firestore-based real-time sync tuned through query optimization; RevenueCat subscriptions; CleverTap-driven analytics and retention campaigns; and a Flutter Web admin panel for the content team.
Architecture
- Flutter app with offline-capable audio playback and background audio
- Firebase: Firestore, Authentication and Cloud Functions as the entire backend
- RevenueCat subscription management; CleverTap analytics + retention tracking
- Flutter Web admin panel for catalog and content operations
Challenges & How I Solved Them
Real-time sync at 100K-user scale on Firestore
Read costs and latency were controlled through query tuning, denormalization and careful listener scoping, the difference between a snappy library and a burning Firebase bill.
Engagement features that moved retention
Audio playback, dark mode and shareable summaries were shipped and measured through CleverTap cohorts, directly improving engagement and retention.
My Contributions
- Built the consumer Flutter app end-to-end across Android and iOS
- Designed and managed the Firebase backend (Firestore, Auth, Cloud Functions)
- Integrated RevenueCat subscriptions and CleverTap analytics
- Developed the Flutter Web admin panel for the content team
Impact
- 100K+ Play Store installs, 4.0+ rating, live on both stores
- Non-technical team manages all content independently via the admin panel
Lessons Learned
Consumer scale on Firebase is an exercise in restraint: model data for reads, scope listeners ruthlessly, and measure every feature against retention.