Inspiration
The idea sparked from a simple thought β what if someone struggling mentally could talk to a supportive AI instantly, without judgment or delay? With rising mental health concerns and lack of accessible support, we felt an urgent need to build something meaningful.
What it does
Itβs a web-based psychologist designed to listen, understand, and support you through thoughtful conversations, self-care tips, and grounding exercises. Whether you're anxious, stressed, or just need to talk β Emosense is there, always. No judgment. Just empathy β backed by AI.β
How we built it
Emosense combines a user-friendly web interface built with HTML, CSS, and JavaScript, powered by Node.js on the backend. It uses the Together.ai API (Mixtral-8x7B-Instruct) to generate emotionally intelligent responses and personalized self-care tips. The chatbot is designed to be empathetic, fast, and context-aware.
Challenges we ran into
Rate limits and inconsistent outputs from LLM APIs Handling user input flow, validation, and real-time updates Designing AI prompts that ensure compassionate and actionable responses
Accomplishments that we're proud of
Built a fully functional mental health chatbot within hours under tight hackathon pressure. Emosense generates emotionally intelligent responses that feel personal β not robotic. Integrated self-care tips, grounding activities, and responsive UI β all with smooth backend AI interaction. Real-time message flow and empathy-first design gave users instant emotional support. Overcame multiple API failures, rate limits, and last-minute UI/UX issues with quick, creative problem-solving.
What we learned
How to integrate and fine-tune AI APIs for real-time user interaction The importance of response filtering to avoid generic, unhelpful replies Optimizing UI/UX for accessibility and comfort during emotional moments
What's next for Emosense β AI Mental Health Support
Mobile App (Flutter) version with emotion-based suggestions and journaling features. Voice interaction and natural language understanding for better accessibility. Integration with mental health professionals for escalated support when needed. Anonymous analytics dashboard to track common mental health trends and improve responses. More focus on privacy, end-to-end message encryption, and user authentication.
Built With
- css3
- dotenv
- express.js
- fetch
- html5
- javascript
- ngrok
- node.js
- together.ai


Log in or sign up for Devpost to join the conversation.