SquadPax, Squad Fitness App
Client, Trenith in-house product
A squad-accountability fitness app built end to end in React Native: AI coach with memory, health-data-aware training, and a store-ready release pipeline.
What was broken
Solo fitness tracking apps lose users fast. SquadPax bets on squad accountability: friends train together, see each other's consistency, and get coached by an AI that knows their actual training history. That demands a real backend, server-enforced privacy, and an AI layer that reads structured data instead of inventing numbers.
What we built
Trenith built the full product in-house: an Expo React Native app on a Supabase backend (Auth, Postgres with row-level security, Realtime, Storage), an AI coach with retrieval-augmented memory fed by logged workouts and daily health activity from Apple Health and Android Health Connect, meal-photo recognition, squad leaderboards, a RevenueCat paywall, and a separate Next.js admin app. Privacy is enforced server-side: an 18+ age gate, scheduled photo purge, and hardened row-level security shipped as numbered migrations.
AI & automation layer
The stack
Outcomes
What this proves
Trenith ships complete mobile products, not prototypes: one React Native codebase, a production backend, an honest privacy layer, and the store pipeline to deliver it.
Sitting with the same problem?
A 60-minute call. You leave with a one-page scope and a fixed number, or a straight "this isn't for us."