Inspiration
InnerGlass AI was built because of our genuine care for the growing mental health crisis among college students. A significant percentage of college students are overwhelmed, anxious, and sometimes depressed. With studies indicating that nearly 36% report that they are not sure what they could do to overcome their feelings. Studies have shown that the overall wellbeing of students significantly improved upon engaging in a few weeks of Cognitive Behavioral Therapy (CBT) and Meditation. Most students are unfortunately not aware of CBT and meditation, and often struggle to afford it. We wanted to create a solution that makes Cognitive Behavioral Therapy and Meditation accessible and judgment-free.
What it does
Our app allows users to express their emotions through speech or text, which offers personalized support using CBT-based reflection, positive reframing, and actionable mental health strategies like journaling and meditation. Each interaction is saved to help users track their growth over time and to observe their emotional patterns.
How we built it
We built the platform using React Native with TypeScript for cross-platform development and used the phone’s local storage for user privacy. Gemini’s API powers the generative responses. We also trained Google Gemini with meditation books and psychological texts to guide CBT-specific support.
Challenges we ran into
We had to ensure that the LLM’s tone maintained empathy such that the users felt as though they were safe to communicate with InnerGlass.
Accomplishments that we're proud of
We’re beyond proud to have developed a functional platform that provides thoughtful and structured emotional guidance. We successfully integrated therapeutic practices into a language model and created an adaptive therapy assistant that customizes its responses for each user!
What we learned
We learned how to find the crossover between psychology and technology. Importantly, our appreciation for empathetic design grew, as well as the importance of feedback loops in improving generative AI. We also became increasingly aware of the responsibility we’re given when building mental health tools.
What's next for InnerGlass AI
We hope to make mental health resources accessible to everybody. Ideally launching a free beta test across college campuses and see the positive impact we made on students. We also aim to add features like mood tracking, cultural context awareness, and perhaps stronger emergency support integration.
Log in or sign up for Devpost to join the conversation.