🌅 Inspiration
Mental health struggles are at an all-time high, especially among students who are juggling classes, social lives, and uncertainty about the future. Journaling can be a powerful outlet — but most people stop after a few weeks because it feels one-sided.
We wanted to make journaling feel more alive with something that listens, understands, and responds. That’s why we built Aurora Journal, a calming space that uses AI to reflect your emotions and help you heal through self-awareness.
đź’ˇ What it does
Aurora Journal is an AI-powered journaling companion that:
- Reads and understands your journal entries in real time
- Detects emotions like joy, anxiety, sadness, and calm
- Generates thoughtful affirmations and reflections
- Visualizes your moods through an interactive “Mood Wheel”
Our goal is to make mental wellness feel approachable and interactive — not clinical or intimidating.
⚙️ How we built it
- Frontend: Flask, HTML, CSS, and JavaScript for a smooth and responsive UI
- Backend: FastAPI for processing journal entries and handling requests
- AI Model: Built using Hugging Face’s
transformerslibrary with emotion classification viadistilroberta-base - Visualization: A dynamic Mood Wheel drawn with JavaScript canvas
- Design: Calming gradients, ambient background music, and minimalist typography
We collaborated on GitHub and used VS Code + virtual environments for efficient teamwork.
🚧 Challenges we ran into
- Integrating FastAPI and Flask into one cohesive system
- Managing merge conflicts while working across multiple branches
- Fine-tuning the AI’s emotion detection for short journal entries
- Styling and synchronizing background music without interrupting the user experience
These challenges helped us improve our debugging, communication, and version-control skills.
🏆 Accomplishments that we're proud of
- Built a fully functioning journaling web app in under 36 hours
- Created an AI feedback loop that feels emotionally intelligent
- Designed a calming, professional-looking UI and logo that matches the project’s mission
- Successfully merged multiple contributors’ work through Git
đź“š What we learned
- How to combine Flask (frontend) and FastAPI (backend) effectively
- Emotion analysis and sentiment modeling using
transformers - How small design choices — fonts, colors, and sound — affect user emotion
- The power of collaboration under time pressure during a hackathon
🌠What's next for Aurora Journal
We plan to expand Aurora Journal with:
- User authentication and private journals
- Weekly mood reports using time-series visualization
- Voice journaling powered by speech recognition
- Mobile compatibility and cloud synchronization
- Integration with smartwatch data to reflect physiological patterns
Aurora Journal is more than a hackathon project — it’s our step toward making mental health reflection accessible, comforting, and beautifully human.
Log in or sign up for Devpost to join the conversation.