How we built it
We built AroMi-AI Agent using React.js with Tailwind CSS for a responsive and intuitive frontend, and Python for backend AI logic and APIs. MongoDB is used to store user profiles, health data, and contextual history. The Groq API powers ultra-fast LLM inference for real-time, adaptive conversations and recommendations.
Challenges we ran into
Integrating Groq API for low-latency, real-time AI responses
Designing a scalable MongoDB schema for evolving user context
Ensuring smooth communication between React frontend and Python backend
Handling health data responsibly while keeping responses empathetic
Balancing UI simplicity with feature-rich wellness insights
What we learned
Fast inference (via Groq) greatly improves user trust and engagement
Context stored in MongoDB enables truly adaptive AI behavior
Tailwind CSS accelerates UI development without sacrificing design quality
AI health systems must prioritize clarity, safety, and explainability
Cross-stack integration is key to building real-time AI products
What’s next for AroMi-AI Agent
Wearable and fitness device integration
Advanced image-based nutrition analysis
Predictive wellness insights using long-term data
Multilingual support and regional health adaptation
AI-generated personalized health reports and coaching plans
Log in or sign up for Devpost to join the conversation.