BrainBloom

AI-powered mental health assistant that helps users navigate emotional challenges with real-time support, peer resources, and crisis detection.

Comment

✨ Inspiration

We were inspired by a rising need for accessible, safe, and supportive mental health check-in spaces, especially ones that don't feel clinical or overwhelming. With so many people struggling to express what they’re feeling, we envisioned BrainBloom: a soft, interactive AI companion that helps users reflect, share, and find calm through simple conversations.

💬 What it does

BrainBloom is an emotionally supportive AI chatbot designed to:

  • Offer a calm space for users to check in on their mental health
  • Engage in empathetic, real-time conversations powered by Gemini
  • Provide tailored responses based on user input and emotion tags
  • Allow users to explore peer support, resources, and personal progress
  • Offer a fully responsive and accessible user interface across all devices

🛠️ How we built it

  • Built with React and TypeScript
  • Styled using Tailwind CSS for a clean and modern UI
  • Framer Motion for subtle animations like Gradient Motion Veil and Breathing Glow
  • Custom screens: Meet the Agents, FAQ, Login
  • Responsive design with mobile-first principles

- Hosted on Google Cloud Platform (GCP)

🧗 Challenges we ran into

  • Managing Gemini’s response tone to maintain emotional safety
  • Implementing layered background animations without UI clutter
  • Balancing performance and aesthetics with subtle effects
  • Ensuring mobile responsiveness while maintaining desktop elegance
  • Structuring frontend and backend communication across secure boundaries

🏆 Accomplishments that we're proud of

  • A soothing and inviting user interface that feels safe and thoughtful
  • Emotion-aware chat flow that encourages self-reflection
  • Fully functional frontend with animated, accessible UI in a short time
  • Achieving emotional tone alignment between design and content

📚 What we learned

  • Emotional UX matters as much as technical UX
  • Subtle design (spacing, softness, colour temperature) drastically impacts emotional perception
  • AI prompt engineering and response control are critical for sensitive domains
  • Working in sync across the frontend and backend tracks accelerates development

🚀 What's next for BrainBloom

  • Personalisation with memory and persistent agent personas
  • A journaling and mood-tracking dashboar
  • AI backend integration
  • Deeper resource integration (e.g., crisis lines, therapy recommendations)
  • Admin panel for content moderation and message tuning
  • Full production deployment with robust analytics and user testing

Built With

Share this project:

Updates