AURA

From daily emotions to burnout prevention — powered by AI.

Comment

Inspiration

Students often experience continuous academic pressure, deadlines, and performance anxiety, but emotional stress usually goes unnoticed until it turns into burnout. While many tools offer chat-based support, few help users understand their emotional patterns over time. This gap inspired us to build AURA — a system that helps students reflect on their emotions, identify stress early, and take small, proactive steps to protect their mental well-being.

What it does

AURA is an AI-powered emotional check-in and burnout early-warning platform for students. Users can express how they feel through text or voice, and the system analyzes emotional tone, sentiment, and key stress indicators. Based on this analysis, AURA generates personalized coping suggestions and short micro-interventions such as breathing or grounding exercises. It also tracks emotional data over time and visualizes trends, helping users recognize patterns in stress and emotional health before burnout escalates.

How we built it

The frontend was built using React with TypeScript, focusing on a clean, responsive, and intuitive user experience. The backend was developed using FastAPI, which exposes REST APIs for processing user inputs and managing emotional data. We integrated Google Gemini to analyze emotional context and generate personalized guidance. Voice inputs are handled on the client side and sent to the backend for processing. The system combines AI-driven analysis with lightweight data storage to support trend visualization and real-time feedback for the MVP.

Challenges we ran into

One major challenge was designing emotional insights that were helpful without being overwhelming or overly generic. Integrating voice and text inputs into a consistent analysis pipeline also required careful handling. Additionally, ensuring fast and reliable AI responses while keeping the MVP simple pushed us to make thoughtful architectural trade-offs.

Accomplishments that we're proud of

We built a fully functional MVP that supports both voice and text-based emotional check-ins, AI-driven emotional insights, and trend visualization. We are especially proud of creating a solution that goes beyond basic chat interfaces by focusing on emotional patterns and early burnout detection. Successfully integrating a modern frontend stack with an AI-powered backend within a limited timeframe was a key achievement for our team.

What we learned

This project helped us deepen our understanding of FastAPI, API design, and frontend development with React and TypeScript. We gained hands-on experience working with generative AI models for real-world problem solving and learned the importance of responsible AI design, especially in sensitive areas like mental health. We also learned how to scope and deliver a meaningful MVP under time constraints.

What's next for AURA

Next, we plan to improve personalization by incorporating longer-term emotional trends and contextual factors like academic schedules. We also aim to add secure user accounts, more advanced analytics, and proactive alerts to help users take action before burnout occurs. In the long term, AURA could expand to support broader student well-being and institutional use cases.

Built With

Share this project:

Updates