Motivio

Motivio lets you enter your personal goals and receive personalized daily motivation to keep you focused and moving forward.

Comment

🚀 Inspiration

Most motivation apps give generic quotes that feel good for five seconds and then get ignored. I built Motivio to be a lightweight AI coach that learns each user's goal and coaching style, then nudges them toward real action. Instead of passive inspiration, every interaction is designed around micro-steps and immediate reflection.

🔎 What It Does

Motivio is a React Native app that delivers personalized daily motivation based on each user's progress.

Core Features:

  • Actionable AI Content: Every generated quote includes a micro-step and reflection question, so users can act right away.
  • Adaptive Coaching: Daily check-ins drive behavior-based coaching. The AI intelligently switches between three modes: reinforce, accountability, and reshape.
  • Smart Scheduling: Quotes are sent at each user's local preferred time via an hourly background dispatcher.
  • Offline-First UX: Users can view history and use the app offline, with a local SQLite cache that syncs to the cloud in the background.

🛠️ How I Built It

This project uses a mobile-first stack with a cloud AI backend:

  • Frontend: React Native + Expo, TypeScript, and Jotai state management.
  • Backend: Firebase Cloud Functions (v2) for scheduled quote generation, adaptive logic, goal validation, and safety checks.
  • AI Engine: Amazon Bedrock (Nova Lite) for structured JSON generation and goal evaluation. The same model also screens new goals for safety.
  • Data Layer: Firestore for cloud profiles/history + Expo SQLite for on-device persistence.

⚡ Challenges

AI Quality Control:
To prevent repetitive outputs, the backend loads recent quote history, runs similarity checks, and retries generation when needed:

similarity > threshold => retry generation (max 3 attempts)

Goal Safety:
The same model is used for both goal quality checks and safety screening to catch harmful or unsafe intent at goal creation.

🏆 Accomplishments

  • Adaptive Behavior State: A trigger-based system (checkinWritten) updates agent state from real-world user consistency.
  • Timezone-aware Dispatcher: scheduledQuoteGeneration reliably finds users whose local time matches their notification hour, worldwide.
  • Secure AI Integration: Transitioned from client-only logic to a full server-side AI pipeline via Bedrock.

📚 What I Learned

This project has been a major bridge in my development journey, especially with Amazon services. As this was my first time using any Amazon service for personal use, I gained significant experience with:

  • Amazon Console & Permissions: Learning to navigate and manage the console, set service permissions, and handle credentials securely.
  • Amazon Bedrock: Developing a deep understanding of Amazon Bedrock, leveraging the Amazon Nova Lite (amazon.nova-lite-v1:0) model for Motivio.

✨ What's Next for Motivio

MVP Deployment & Feedback
I plan to deploy the MVP and gather real-world user feedback. This will help guide improvements and refinements based on actual usage.

Improved Quote Generation
I'll keep refining model output. Adaptive modes (reinforce, accountability, reshape) will provide deeper, more relevant insights as the user progresses.

Smart Notifications
To make notifications "smart", I plan to use Amazon Nova Lite to reason about each user's specific goal. The model will determine optimal content and timing, rather than relying on a fixed user setting.

Voice Motivation with Amazon Polly
I aim to integrate Amazon Polly as a Text-to-Speech (TTS) feature. Motivation via voice is often more impactful than text, so users will be able to hear their daily quotes.

Built With

Share this project:

Updates