About the Project: SerenityAI
🎯 Inspiration
Mental health care is often inaccessible due to stigma, cost, and availability. The inspiration for SerenityAI came from a deeply personal place—witnessing friends and family struggle to find timely, affordable mental health support during the COVID-19 pandemic. The isolation and anxiety of those times highlighted a critical gap in accessible mental health resources.
I envisioned a platform where anyone, anywhere, could receive immediate, stigma-free support. By blending AI-driven tools with human connection, SerenityAI aims to make mental wellness accessible to all, anytime—breaking down barriers that too often stand in the way of care.
📚 What I Learned
Building SerenityAI was a journey of growth, both technically and personally. Key areas of learning include:
🤖 AI Integration
- Developed an empathetic AI chatbot using Bolt.new
- Focused on natural language processing for mood tracking and crisis detection
- Fine-tuned the model to detect emotional nuances—e.g., distinguishing a bad day from a true crisis
💳 Payment Systems
- Integrated Stripe for therapist payments
- Implemented custom pricing and free trials
- Balanced flexibility and security for a seamless transaction experience
🔐 Compliance
- Ensured HIPAA-compliant data handling
- Gained experience with encryption and access controls
- Designed systems that protect privacy while enabling real-time crisis response
🛠️ How I Built It
SerenityAI was crafted for scalability, security, and user experience. Here's a breakdown of the tech stack and features:
🧠 Bolt.new
- Core app and AI chatbot built with Bolt.new
- Enabled quick iterations on chatbot responses
- Provided meaningful and supportive user interactions
🗃️ Supabase
- Backend with real-time database for:
- User authentication
- Mood logs
- Therapist profiles
- Enabled scalable data handling and instant crisis alerts
💸 Stripe Connect
- Allowed therapists to set custom pricing
- Enabled direct payouts and free trials
- Integrated seamlessly with Supabase for unified user/payment management
📞 ElevenLabs & Twilio
- Built the “Call Me” feature:
- ElevenLabs for natural TTS
- Twilio for call handling
- Focused on low latency and voice quality for a comforting experience
🚧 Challenges Faced
Developing SerenityAI brought its share of challenges:
🔌 API Integration
- Syncing Stripe, Supabase, and Twilio was complex
- Required custom webhook logic to handle real-time updates between systems
🛡️ Compliance
- Balancing privacy and crisis detection within HIPAA regulations
- Implemented granular access controls for sensitive data like mood logs and chat history
🎨 User Experience
- Designed for both patients and therapists
- Prioritized accessibility and ease-of-use
- Conducted user testing and met WCAG 2.1 standards for inclusivity
Testing Credentials
- Email:
therapist@gmail.comto login as Therapist - Email:
patient@gmail.com- to login as Patient
- All test accounts use the password: Patient@123
🌈 Final Thoughts
In the end, SerenityAI became a platform that combines the best of AI and human care, delivering mental health support at your fingertips. By addressing accessibility with innovation, it’s a step toward a world where mental wellness is within reach for everyone.
Built With
- bolt.new
- elevenlabs
- javascript
- node.js
- openai
- postgresql
- react
- stripe
- supabase
- twilio
Log in or sign up for Devpost to join the conversation.