HowUBeen

Inspiration

This project was inspired by the app 活着么, a safety-focused application that requires users to check in daily to confirm they are still alive, with missed check-ins triggering emergency contact notifications. While we found this idea powerful, we felt it addressed only physical safety. We wanted to expand this concept into a broader mental health and wellness tool — one that ensures users are not only alive, but present, supported, and staying on track with their lives. In a world where many people struggle to maintain goals and routines, especially when living alone, we wanted to create a system of gentle accountability rather than pressure or guilt.

What it does

HowUBeen is a mental health and wellness app centered around simple daily check-ins. Users start with a quick set up where they include there own name, email, and then the emails of their emergency contacts, and then includes 3 big focuses of there life they have for a set next period of time. After that, every day users can quickly(under 1 minute) check in using either text or voice, sharing how they’re feeling and how aligned they are with their personal goals. These check-ins are analyzed using AI to extract key insights such as mood, focus, and overall well-being. The system generates weekly summaries to help users reflect on patterns over time. If a user misses a check-in or if concerning signals are detected, the app automatically notifies a designated emergency contact, enabling early support without requiring the user to explicitly ask for help.

How we built it

We built HowUBeen using a React-based frontend and a Node.js backend powered by Express.js. User data and check-ins are stored in a lightweight SQLite database. To keep the experience seamless, we integrated AI services to handle both text and voice inputs. Voice check-ins are transcribed using OpenAI Whisper, while GPT-4o-mini analyzes the content to extract mood, goal alignment, and key takeaways. Background jobs handle inactivity checks and trigger emergency email notifications when needed.

Challenges we ran into

One of our biggest challenges was integrating multiple external APIs, particularly the OpenAI services for AI analysis and the email system for emergency notifications. Ensuring these services worked reliably together within tight hackathon time constraints was difficult. Additionally, organizing the project structure and maintaining clarity between frontend, backend, and AI logic required careful planning and constant refactoring.

Accomplishments that we're proud of

We are especially proud of the concept behind HowUBeen. We believe the idea of combining safety, mental health, and gentle accountability in a single streamlined experience has real-world impact. Despite limited time, we successfully built a working end-to-end system that demonstrates how technology can support both well-being and personal responsibility without inducing guilt or pressure.

What we learned

• How to design an MVP by prioritizing core functionality over polish
• Integrating AI services meaningfully rather than as a novelty
• Working with asynchronous APIs and background jobs
• Structuring a full-stack project for scalability and clarity
• Collaborating effectively under tight time constraints ## What's next for HowUBeen What we made here is the most minimum viable project, we have much more to do. The ultimate goal is to deploy it because it will have real world effects and designed as a practical and effective solution. Next, we want to expand HowUBeen into a fully accessible mobile experience, improve AI personalization over time, and add smarter detection for long-term patterns in well-being. We also plan to explore additional notification methods and privacy-preserving features to ensure users feel safe, supported, and in control of their data.

Built With

• 18
• 20.11.0
• api
• axios
• css3
• email
• es6+)
• ethereal
• express.js
• gpt-4o-mini
• javascript
• multer
• node-cron
• node.js
• nodemailer
• openai
• react
• sqlite
• vite
• whisper