Existing sports coaching solutions (TrueCoach, Trainerize, Everfit) are English-first with approximate French translations. Francophone gym coaches waste time navigating interfaces designed in English. On the athlete side, the mobile experience is often broken: no integrated readiness check, no auto-detected PRs, no native rest timer. Engagement drops after 30 days in 80% of market solutions.

Native mobile bilaterality: both coach AND athlete must have a complete mobile experience, no "use desktop for programming". Native French, not an i18n filter applied after the fact. UI designed in Quebec French. Non-negotiable security from Sprint 1: health data (readiness, injuries, weight) subject to Loi 25 + PIPEDA + GDPR-ready. Bootstrap with no external funding. Strict beachhead: 50 paying coaches + NPS > 40 + M3 retention > 70% before expanding.

Mobile-first architecture with Expo SDK 52+ (React Native), strict TypeScript and NativeWind. Supabase for auth, Postgres database and RLS on 100% of tables from Sprint 1. Next.js landing page with active waitlist. Exhaustive competitive audit of 8+ apps (TrueCoach, Trainerize, TrainHeroic, Everfit, Hevy, CoachNow, TeamBuildr, PT Distinction) to identify gaps. 169+ structured GitHub issues with detailed acceptance criteria and screen-by-screen user flows. Alexandre (design co-founder) always works 1 sprint ahead on Figma mockups. User testing planned at Sprints 3, 5, 7, 9.

169+ GitHub issues structured across 2 repos (41 cadence-mobile + 128+ the-project). 13 sprints planned with rigorous versioning (alpha → beta → RC → prod). 74/74 market features covered after audit (100% coverage across 7 functional pillars). 36 feature issues with complete screen-by-screen user flows. 10 foundational principles documented as decision filters. Security by design: Sprint 1 RLS, AES-256 encryption at-rest, TLS 1.3, short-lived JWTs, OWASP Mobile Top 10 audit planned pre-beta.

Planning as a Tech Lead changes everything. Doing 90% of planning BEFORE code means fewer surprises, clearer scope and avoided tech debt. The discipline of "if it's not MVP, it's later" reduced scope from ~40 features to ~25 surgical ones. Having the design co-founder one sprint ahead eliminates blockers: Figma mockups are ready when dev arrives. The 10 foundational principles (Athlete-First, Ship > Perfect, Data > Opinions) serve as decision filters that prevent unproductive debates.