MindMesh

Inspiration

Mental health is a growing concern worldwide, and many people struggle to track their emotional well-being or reach out for help in times of crisis. I wanted to create a tool that not only helps users understand their feelings but also provides real-time support and emergency intervention when needed.

What it does

MindMesh is an AI-powered emotional wellness companion. Users can check in with their feelings, and the app uses advanced language models (LLMs) to analyze mood, provide personalized insights, and track emotional trends. If a user expresses a crisis (e.g., “I want to die”), MindMesh instantly detects it, displays a crisis alert with resources, and notifies emergency contacts via email.

How we built it

Frontend: Built with React, Vite, and Tailwind CSS for a fast, responsive, and modern UI. AI Integration: Uses OpenAI’s GPT-4o-mini (or other LLMs) for deep emotional analysis and chatbot responses. Crisis Detection: Custom logic to identify crisis phrases and trigger emergency workflows. Notifications: Integrates with EmailJS to send real-time alerts to emergency contacts. Persistence: All user data (mood history, analytics, personal space) is saved in localStorage for privacy and offline access. Analytics: Visualizes mood trends, streaks, and emotional heatmaps to help users track their progress.

Challenges we ran into

Ensuring reliable crisis detection and not missing any critical messages. Handling email delivery and third-party service configuration. Making the app intuitive and supportive for users in distress. Persisting user data securely and restoring it after login or refresh.

Accomplishments that we're proud of

What we learned

How to integrate advanced AI models into a real-world app. The importance of user data privacy and persistence. Building robust crisis detection and notification systems. Creating a seamless experience across web and mobile.