MindEase

Inspiration

Journaling is a proven way to improve mental health, but a blank page can feel intimidating. When someone is experiencing distress, writing alone isn't always enough. We wanted to build a supportive tool that combines AI empathy with gamification to make taking care of your mental health feel highly rewarding.

What it does

MindEase is a highly interactive, gamified AI journaling companion.

• Virtual Economy: Users earn "Mind Coins" for building healthy habits.
• AI Mood Analysis: Google Gemini analyzes text to detect your mood and provide a warm reflection.
• Crisis Safety Layer: A deterministic algorithm ensures user safety during distress.

To see our UI in action, check out our live demo deployment on Vercel (or view the screenshots below).

How we built it

We built the frontend using React and Vite. The backend is built with Node.js and Express.

One of the coolest features is our Sprout Evolution Logic. The virtual companion levels up logarithmically based on the coins earned. The level ( L ) is calculated using this in-line math formula: ( L = \lfloor \frac{\text{Coins}}{50} \rfloor + 1 ).

If a user earns coins steadily over time, the total experience curve required to reach maximum level is defined by the following displayed equation: $$ E_{total} = \sum_{n=1}^{5} (n \times 50) $$

Challenges we ran into

Ensuring safety was difficult. Large Language Models can hallucinate, so we couldn't rely on them for crisis detection. We built a deterministic safety layer that intercepts requests before they reach the AI.

Here is a simplified code block showing how our deterministic safety check works:

export function detectCrisis(text) {
  const crisisKeywords = ["give up", "can't go on", "end it", "hurt myself"];
  const lowerText = text.toLowerCase();

  for (const word of crisisKeywords) {
    if (lowerText.includes(word)) {
      return true; // Instantly trigger emergency resources
    }
  }
  return false;
}

Accomplishments that we're proud of

We are incredibly proud of successfully taking a serious topic—mental health tracking—and turning it into a delightful, rewarding experience.

What we learned

We learned advanced prompt-engineering techniques to ensure the Gemini model responded with empathy and validation. We also learned how to build tactile 3D CSS animations to make the interface feel "alive."

What's next for MindEase

We plan to introduce full User Accounts with encrypted cloud storage, voice-to-text journaling, and personalized, AI-generated weekly wellbeing summaries to help users spot long-term emotional patterns.

Built With

• backend
• built-using-react-and-vite-for-the-frontend-ui
• hosted
• integrating-the-google-gemini-api-for-core-ai-mood-analysis-and-multimodal-image-transcription.-the-frontend-is-hosted-on-vercel
• is
• on
• render
• styled-with-custom-gamified-css3-and-visualized-using-chart.js.-the-backend-is-powered-by-node.js-and-express.js