Mentora_ai

Your AI study companion that cares

Comment

INSPIRATION

As students ourselves, we've experienced the overwhelming stress of balancing academic performance with mental health. Traditional tutoring focuses only on grades, while mental health resources are often separate and hard to access. We were inspired to create Mentora - an AI companion that doesn't just help you learn, but truly cares about your overall wellbeing. We envisioned a world where every student has a personalized mentor available 24/7, one that adapts to both their learning style and emotional needs.

WHAT IT DOES

Mentora is a dual-purpose AI platform that serves as both an intelligent tutor and a mental health companion. Students can:

  1. Learn with AI tutoring: Get personalized lessons, visual explanations, and practice tests on any topic
  2. Care for mental health: Access stress management, mood tracking, and emotional support
  3. Customize their experience: Name their AI mentors and choose personalities that resonate with them
  4. Track progress: Monitor learning streaks, test scores, and wellbeing check-ins
  5. Switch seamlessly: Toggle between study mode and wellbeing mode based on immediate needs

The platform recognizes that academic success and mental wellness are deeply interconnected, providing holistic support for the modern student.

HOW WE BUILT IT

We built Mentora using modern web technologies for maximum accessibility:

  1. Frontend: React.js with Tailwind CSS for a responsive, mobile-first design
  2. Backend: Node.js with Express.js handling API routes and user management
  3. AI Integration: OpenAI GPT-4 API with carefully crafted prompts for tutoring and wellbeing contexts
  4. Authentication: Google OAuth for seamless user onboarding
  5. Database: MongoDB for storing user profiles, conversations, and progress data
  6. Real-time Features: Socket.io for live chat functionality and typing indicators
  7. Deployment: Hosted on Replit for quick iteration and demo accessibility

We implemented separate conversation histories and system prompts for each AI mode, ensuring contextually appropriate responses whether students need academic help or emotional support.

CHALLENGES WE RAN INTO

  1. API Rate Limiting: Optimizing OpenAI API calls while maintaining conversation quality required implementing smart caching and response chunking
  2. Context Switching: Creating seamless transitions between tutor and wellbeing modes without losing conversation context proved technically challenging
  3. Responsive Design: Ensuring the chat interface worked flawlessly across all device sizes, especially on mobile where students often study
  4. AI Prompt Engineering: Crafting prompts that generate educational content while maintaining appropriate boundaries for mental health support
  5. Time Constraints: Balancing feature completeness with code quality in the hackathon timeframe required tough prioritization decisions

ACCOMPLISHMENTS THAT WE'RE PROUD OF

  1. Holistic Approach: Successfully integrated academic tutoring with mental health support in a single, cohesive platform
  2. User Personalization: Implemented meaningful customization that makes each student's AI feel uniquely theirs
  3. Responsive AI: Created context-aware AI responses that adapt tone and content based on whether students need academic or emotional support
  4. Intuitive UX: Designed an interface so simple that students can focus on learning rather than navigating technology
  5. Real Impact Potential: Built something that addresses genuine pain points we've experienced as students
  6. Technical Excellence: Delivered a fully functional, scalable application with clean code and modern architecture

WHAT WE LEARNED

  1. AI Integration: Mastered prompt engineering techniques for educational and therapeutic contexts
  2. Full-Stack Development: Gained deeper experience with React-Node.js integration and real-time web applications
  3. User-Centered Design: Learned the importance of designing for emotional states, not just functional requirements
  4. API Management: Developed skills in rate limiting, error handling, and optimizing external API usage
  5. Rapid Prototyping: Improved our ability to quickly validate ideas and iterate based on testing feedback
  6. Team Collaboration: Enhanced our skills in dividing complex features across team members efficiently.

WHAT'S NEXT FOR MENTORA

  1. Advanced Analytics: Implement learning pattern recognition to provide personalized study recommendations and early intervention for mental health concerns
  2. Collaborative Features: Add study groups, peer support networks, and teacher dashboards for comprehensive educational ecosystems
  3. Mobile App: Develop native iOS and Android apps with offline capabilities and push notifications for study reminders
  4. Integration Ecosystem: Connect with popular student tools like Google Classroom, Canvas, and calendar applications
  5. Specialized AI Models: Train domain-specific models for subjects like advanced mathematics, coding, and creative writing
  6. Accessibility Enhancement: Add voice interaction, multiple language support, and features for students with learning differences
  7. Institutional Partnerships: Scale to work with schools and universities as an official student support tool
  8. Crisis Prevention: Implement advanced mental health screening with professional counselor referral systems

Mentora represents the future of education - where technology doesn't just make us smarter, but helps us become more resilient, confident, and emotionally intelligent learners.

Built With

Share this project:

Updates