• 

  CuraSense Landing Page

🧠 Inspiration

In today’s fast-paced world, physical health is tracked constantly — but emotional and mental well-being often go unnoticed. We were inspired to build CuraSense to bridge this gap using the power of AI. What if your device could understand how you feel, not just how many steps you take?

Our goal: create a personal wellness assistant that listens to your tone, observes your expressions, and gives you meaningful feedback — just like a real wellness coach.

💡 What it does

CuraSense is an AI-powered, multi-agent health companion that provides real-time emotional insights using voice and facial analysis.

Key Features:

• 🎙️ VoiceAgent: Detects stress, energy, and mood through tone, pitch, and speech patterns
• 👁️ VisionAgent: Analyzes facial expressions to detect emotions like fatigue, anxiety, or happiness
• 🧠 InsightAgent: Combines voice + vision data to generate a unified wellness report
• ⚡ Smart Scan Mode: One tap runs all enabled agents in parallel and presents actionable insights
• 📊 Dashboard: Beautiful, animated UI with real-time metrics and agent logs
• 🔧 Manual Agent Control: Users can enable/disable agents individually in the Agent Center

🏗️ How we built it

We built CuraSense using a modular, event-driven architecture:

• 🧠 Agent Development Kit (ADK) for building and orchestrating autonomous AI agents
• ☁️ Google Cloud:
  • Cloud Run – to deploy and scale agents
  • Cloud Speech-to-Text and Vision APIs – for real-time input analysis
  • BigQuery – for wellness trends over time
• 💻 React + Tailwind CSS – to build an elegant, responsive frontend
• 🔄 Framer Motion & Lottie – for high-end animations and micro-interactions
• 🔐 Firebase Auth & Realtime DB – for agent logs and secure user flow
• 🧩 Figma, IconScout, StorySet – to create cohesive visual storytelling and UX

🚧 Challenges we ran into

• 🤖 Synchronizing multiple agents and their data pipelines in real-time
• 🎯 Combining outputs into a meaningful summary using the InsightAgent logic
• ⚙️ Tuning the vision/voice models to reduce false positives
• 💅 Designing a polished, intuitive interface while maintaining performance
• 🔄 Ensuring modularity so that agents could work independently or collaboratively

🏆 Accomplishments that we're proud of

• ✅ Successfully integrated and orchestrated a multi-agent wellness system
• ✅ Built a fully functional Smart Scan workflow with real-time collaboration between agents
• ✅ Designed a clean, beautiful, and animated dashboard that enhances user experience
• ✅ Created a unique insight engine that contextualizes raw agent data into actionable feedback

📚 What we learned

• The power of agent-to-agent (A2A) communication for scalable AI systems
• The importance of UI/UX in health-tech — clarity and trust matter as much as features
• Hands-on experience with Google Cloud services, especially Cloud Run and BigQuery
• Building a real-time pipeline that combines audio, video, and AI models on the edge and cloud

🔮 What's next for CuraSense

• 📱 Launching a mobile-first version using Flutter
• 🤝 Adding integration with wearables (e.g., Fitbit, Apple Health)
• 📦 Creating a customizable agent SDK for healthcare providers and therapists
• 🧪 Introducing mood journaling + AI recommendations based on daily patterns
• 🌍 Expanding to multilingual, culturally adaptive emotional models

Built With

• api
• bigquery
• cloud
• cloudrun
• firebase
• google
• react