MindBridge AI

Inspiration

1 in 4 people worldwide suffer from mental illness, yet over 80% lack access to proper care. In India, there are only 0.3 psychiatrists per 100,000 people. We wanted to bridge this gap using AI — giving everyone access to a judgment-free, always-available mental health companion.

What it does

MindBridge AI is a web platform with three core features:

• AI Chat Companion: Real-time empathetic conversations powered by Llama 3.3 70B via Groq API
• Mood Tracker: Daily mood logging with visual analytics to identify emotional patterns
• Resource Hub: Curated mental health resources and verified Indian crisis helplines (iCall, Vandrevala Foundation)

How we built it

Built with React + Vite for a fast frontend, TailwindCSS for responsive design, Groq API (Llama 3.3 70B) for real-time AI chat, and Recharts for mood visualization. The app is 100% anonymous — no sign-up, no data storage.

Challenges we ran into

• Groq deprecated the llama3-8b-8192 model mid-development, requiring a quick switch to llama-3.3-70b-versatile
• Designing a UI that felt warm and safe, not clinical
• Balancing AI helpfulness with responsible mental health boundaries

Accomplishments that we're proud of

• Built a fully functional AI mental health platform in under 5 hours
• Integrated real-time AI with full conversation memory
• Designed a premium, accessible UI with crisis helplines always visible
• 100% free and anonymous — zero barriers to access

What we learned

• How to integrate Groq's LLM API for real-time conversations
• Importance of responsible AI design in sensitive domains
• How to rapidly prototype and iterate under time pressure

What's next for MindBridge AI

• Multi-language support (Hindi, Tamil, Telugu)
• Integration with NGOs and healthcare providers
• Mobile app version
• AI-powered journaling and CBT exercises

Built With

• css3
• framer-motion
• groq
• html5
• javascript
• llama-3.3-70b
• note.js
• react
• recharts
• tailwindcss
• vite