Problem
What was needed?
"People want to be heard, not necessarily seen." The goal was a calm, voice-first social space outside the visual-feed paradigm.
Case study
MindRooms
A social podcast platform where voice comes before visuals — acoustic identity, live rooms, and community discovery in one app.
Platforms
iOS / Android
Role
End-to-end mobile development, Supabase architecture, and design-verification tooling.
Tech stack
Flutter · Riverpod · Supabase
Status
Launch-ready
Solution
How was it solved?
A 44-screen Figma design was translated to code with a custom pixel-diff verification pipeline (~2% average deviation). Supabase powers 9 tables, 27 row-level security policies, and Edge Functions behind auth, discovery, the player, and like/comment/follow flows.
Outcomes
What came out of it?
44/44 screens at ~2% average pixel deviation from design
Voice signature: acoustic identity captured at signup, stored privately
Background audio playback with lock-screen controls
Username login with server-side mapping that never leaks emails
Build note
Build narrative
MindRooms treated design fidelity as a measurable engineering target. A custom toolchain pulls screen specs from the Figma REST API, captures simulator output, and quantifies the difference — every screen shipped with a verified score. On the product side, live podcast rooms, category discovery, communities, and follows sit on top of Riverpod state management, secure session storage, and row-level authorization.
Product experience
Client app and management panel
One tracking system, delivered through two interfaces shaped around each role’s daily workflow.
Guided onboarding
Username sign-in
Live podcast room
Category discovery
Profile and follows
Voice signature capture