Inspiration
PocketCoach was inspired by Simon’s vision of making high-quality coaching accessible, simple, and available exactly when it’s needed. Coaching has an enormous impact on personal growth, but for most people it’s still expensive, time-consuming, and hard to access. At the same time, modern AI agents are powerful but often feel technical and overwhelming. PocketCoach bridges that gap with a minimalist, human-centered AI coaching app.
What it does
PocketCoach lets users browse or create AI coaches, add personal context like goals and values, and instantly start chatting with a coach for guidance and clarity.
Users can share coaches, personalize their experience, and unlock premium features like unlimited messages and custom coaches via subscription.
How we built it
PocketCoach is built as a cross-platform mobile app using Flutter.
Personalized coaching is powered by an LLM API, combining the selected coach’s personality with the user’s personal context.
Subscriptions, paywalls, and entitlements are handled using RevenueCat.
Challenges we ran into
The main challenges were designing prompts that feel genuinely “coach-like” instead of generic, and creating a UX that hides AI complexity and feels calm, simple, and human.
Accomplishments that we're proud of
- A polished, minimalist mobile experience
- A working AI coaching flow with real personalization
- A production-ready subscription system using RevenueCat
- A clear, creator-aligned MVP that feels like a real product
What we learned
We learned how sensitive AI behavior is to prompt design, how important UX is for perceived intelligence, and how much time RevenueCat saves when building subscription products.
What's next for PocketCoach
Next steps include voice input, scheduled coaching nudges, deeper personalization, and integration with other apps.