# Project export: Solarii

This document was generated by HackStack to give an AI agent context about a hackathon project. Sections are labeled with their provenance; content marked as truncated was cut to keep this document small.

## Project metadata

- Hackathon: UC Berkeley AI Hackathon 2025
- Tagline: Solarii is a trauma-aware journaling web app that uses Claude, Gemini, and Vapi to help users reflect, heal, and grow.
- Devpost: https://devpost.com/software/solarii
- GitHub: not linked
- Team: contributor stats unavailable

## Devpost submission (written by the team)

### Inspiration

Mental health support is often inaccessible, expensive, or stigmatized. We were inspired to create a safe, private space where individuals can process emotions, track mental wellness patterns, and access personalized AI therapy support 24/7. Solarii bridges the gap between traditional journaling and professional mental health resources.

### What it does

Solarii is an intelligent mental wellness platform that combines emotional journaling with AI-powered therapeutic support. Users can: Journal with emotion tracking - Write entries and tag emotional states for pattern recognition Access 4 specialized AI therapists - Each with unique therapeutic approaches (CBT, mindfulness, holistic wellness, peer support) Visualize emotional patterns - Interactive emotional mapping and trend analysis Generate AI reflections - Personalized insights based on journal entries Voice-to-text journaling - Seamless audio transcription for accessibility

### How we built it

Frontend: Next.js 14 with TypeScript, Tailwind CSS, and shadcn/ui components for a responsive, accessible interface AI Integration: Multi-provider approach using: Google Gemini Pro for CBT therapy responses Groq Llama-3.1-70B for mindfulness coaching Vetti AI for peer support conversations Claude for holistic wellness guidance Backend: Next.js API routes with Supabase for real-time data storage and user management Key Features: Real-time conversation storage, emotional pattern analysis, responsive design, and graceful API fallbacks

### Challenges we ran into

Multi-AI provider coordination - Integrating 4 different AI APIs with unique authentication and response formats Emotional data visualization - Creating meaningful, non-overwhelming representations of complex emotional patterns Therapeutic boundaries - Ensuring AI responses maintain appropriate professional boundaries while being helpful Real-time performance - Optimizing response times across multiple AI providers while maintaining conversation context

### Accomplishments we're proud of

Seamless AI integration - Successfully orchestrated 4 different AI providers in one cohesive experience Therapeutic authenticity - Each AI agent has distinct personalities and evidence-based therapeutic approaches Accessibility focus - Voice transcription, responsive design, and screen reader compatibility Data privacy - Secure, encrypted storage with user control over their mental health data Scalable architecture - Clean separation of concerns allowing easy addition of new AI providers or features

### What we learned

AI personality design - How to craft distinct, therapeutically appropriate AI personas Mental health UX - The importance of calming design, clear navigation, and crisis support resources Multi-provider resilience - Building robust fallback systems for AI service interruptions Emotional data handling - Sensitive approaches to storing and visualizing personal mental health information

### What's next

Advanced analytics - Predictive emotional pattern recognition and personalized wellness recommendations Community features - Anonymous peer support groups and shared coping strategies Integration expansion - Wearable device data, calendar integration, and mood-based music recommendations Professional partnerships - Connecting users with licensed therapists when AI support isn't sufficient Mobile app - Native iOS/Android applications for on-the-go mental wellness support

## README (from the GitHub repository)

No README available.

## Detected evidence (automated analysis)

No repository was indexed for this project. Claimed technologies below could not be checked against code.

## Codebase structure

No repository index available.

## Key source files

No repository index available; no source files included.