GotItLife AI-Powered Mental Health App

GotItLife is an AI-powered mental health and self-coaching app designed to help people improve psychological resilience, reduce stress and anxiety, and increase focus and productivity through guided internal dialogue and AI coaching. It’s positioned as a tool for personal growth and mental wellbeing, and also marketed to employers as a way to support workforce wellness and performance.

Tech Stack: React Native, Expo, NativeWind, Zustand, TanStack Query, Supabase, PostgreSQL, Open AI, ElevenLabs

I contributed to the development of GotItLife, a mobile AI-powered mental health and self-coaching app designed to help users reduce stress, improve focus, and build psychological resilience through natural language conversations.

I implemented a scalable React Native + Expo frontend, leveraging NativeWind for responsive styling, Reanimated for smooth animations, and Zustand for structured state management to deliver a fast, calm, and interactive chat experience. Conversation threads, reflection prompts, and mood tracking were organized to ensure clarity and engagement while supporting accessibility and soft, psychologically safe UI design.

We integrated large language models (OpenAI GPT-4/5) with a prompt-engineered coaching architecture, allowing AI to interpret user input while following controlled guidance to prevent hallucinations or harmful suggestions. Streaming AI responses were implemented token-by-token for a conversational, real-time experience, and session history was stored locally and synced to a backend to maintain context across interactions.

The app architecture also included offline caching of conversations, secure storage of sensitive data via Expo SecureStore, and push notifications for reminders and progress tracking. Analytics were integrated using PostHog to monitor user engagement, and subscription management was handled via RevenueCat.

AI sometimes gives advice that isn’t accurate or safe. We handled this by carefully crafting prompts, adding rules to catch risky outputs, and checking responses on the backend.

Keeping track of conversations across sessions was tricky. We solved this by saving chat history both locally and on the server, using AI memory techniques to maintain context.

Streaming AI responses in real time could lag or feel slow. We fixed this by sending responses token by token and updating the UI asynchronously so it felt smooth and natural.

Users’ mental health data is sensitive. We made sure everything was stored securely, encrypted during transmission, and compliant with GDPR/HIPAA rules.

Making the app look and work the same on both iOS and Android required careful testing. We built reusable components and tested extensively to ensure a consistent experience

Results: The app delivers a smooth, interactive, and safe AI-driven coaching experience, empowering users to self-reflect, track progress, and improve mental wellbeing while maintaining data security and high responsiveness.