Inspiration
Young people today navigate academic pressure, emotional ups and downs, burnout, and stress—often without a safe space to express what they feel. I wanted to build something warm, accessible, and helpful: an AI companion that supports students with emotional clarity, reflection, and daily wellness habits.
What it does
YouthWell is an AI-powered wellness companion that helps users:
- Track their daily mood
- Reflect through mindful journaling
- Challenge negative thoughts using CBT-style guidance
- Chat with a supportive AI buddy for emotional clarity
- Take standardized tools like PHQ-9 to understand mental health patterns
- Build a healthy wellness routine
All features run on a lightweight architecture that works smoothly on ARM-based devices (Android phones, tablets, and mobile browsers).
How I built it
- Designed a multi-feature wellness interface using Streamlit
- Implemented NLP-based emotional support using lightweight models
- Added mood tracking, journaling, CBT records, and PHQ-9 logic
- Connected it with a lightweight backend suitable for mobile devices
- Ensured the system remains fast and efficient on ARM processors
- Created a clean and calming UI to make wellness feel approachable
Challenges
- Making AI responses feel empathetic instead of robotic
- Keeping the app lightweight enough for mobile/ARM execution
- Designing multiple wellness tools while keeping the flow simple
- Maintaining a balance between mental health seriousness and ease of use
Accomplishments
- Built a complete wellness assistant with multiple real features
- Created an interface that feels encouraging and judgment-free
- Integrated several mental-health techniques (mood tracking, journaling, CBT, PHQ-9)
- Made the system ARM-friendly and mobile-accessible
What’s next
- Personalized wellness plans based on user patterns
- Voice-based emotional check-ins
- Sentiment analysis for journaling entries
- Daily progress streaks and recommendations
- A mobile app version for offline use
Built With
- arm
- mobile-ai
- natural-language-processing
- python
- streamlit
- tensorflow-lite
Log in or sign up for Devpost to join the conversation.