Inspiration
Students often struggle with stress, anxiety, and academic pressure but hesitate to seek help. We wanted to build a private, stigma-free space where anyone can express their feelings and receive supportive guidance. This led to the creation of SafeSpace.ai, an empathetic mental-health companion powered by AI.
What it does
SafeSpace.ai provides:
AI emotional-support chat
Sentiment + emotion detection
Personalized coping strategies
Crisis-risk identification
Daily mood tracking
It creates a safe, judgment-free environment for students to manage their mental well-being.
How we built it
Developed an NLP pipeline for emotion and sentiment classification.
Designed an AI support engine using transformer models and guided response rules.
Built a web interface with React + Node.
Used a secure database for anonymized mood logs.
Added encryption and privacy-first architecture to protect users.
Challenges we faced
Ensuring the AI remains empathetic and safe during sensitive conversations.
Balancing personalization with privacy—no personal user data stored.
Achieving accurate crisis detection without false positives.
Integrating psychological methods like CBT into AI responses.
What we learned
How to design AI that is not only intelligent but also emotionally aware.
Responsible use of NLP in mental-health applications.
Importance of user testing for tuning model accuracy.
Ethical considerations in data handling and conversational AI.
What’s next
Multi-language support
Mobile app version
Voice-based emotion analysis
Gamified wellness features
Collaborations with mental-health experts
Log in or sign up for Devpost to join the conversation.