MindScribe AI

Inspiration

Many individuals silently battle stress, anxiety, or emotional overwhelm. Talking to someone isn’t always easy. We wanted to build a tool that lets users express their emotions freely, privately, and still receive helpful feedback — hence, MindScribe AI was born.

What it does

MindScribe AI is a journaling assistant powered by emotional AI. Users write their thoughts, and the system analyzes them using NLP to detect emotions and sentiment. It then responds with personalized reflections, wellness tips, and mental health prompts — encouraging mindfulness and self-awareness.

How we built it

We used React.js for the user interface, Python (Flask) for the backend API, and Hugging Face Transformers (like BERT) for sentiment/emotion analysis. User authentication and data were managed through Firebase. Deployment was done on Render/Vercel.

Challenges we ran into

• Making the AI’s tone feel supportive and non-robotic
  
• Ensuring complete user privacy and journaling data security
  
• Balancing mental wellness assistance without mimicking therapy
  
• Managing NLP model accuracy for diverse emotional expressions ## Accomplishments that we're proud of
  
• Developed a working MVP that detects emotion from text and responds meaningfully
  
• Created a clean, user-friendly journaling interface
  
• Addressed an important but often ignored problem — digital mental health ## What we learned
  
• Practical implementation of NLP emotion models
  
• UI/UX design for emotionally sensitive applications
  
• Importance of trust, privacy, and safety in AI mental health tools
  
• How to work fast and build something impactful during a hackathon

Built With

• css
• frontend:-html
• javascript-(react.js)-backend:-python-(flask-or-fastapi)-ai/nlp:-hugging-face-transformers-(bert-for-sentiment/emotion-analysis)
• openai-api-database:-firebase-or-mongodb-hosting:-github-pages-+-render-/-vercel-others:-google-auth-/-firebase-auth
• tailwind