AURA — AI-Powered Personal Wellbeing Twin

• 

  Welcome screen where users enter their email address to continue into the AURA wellbeing experience.

• 

  Four-digit verification-code screen used to confirm the user’s identity before continuing.

• 

  Onboarding screen where users choose goals

• 

  User selects their typical time of day, available time, and preferred type of wellbeing support.

• 

  Confirmation screen summarizing the user’s primary goal, routine, preferred duration, and preferred support style.

• 

  Personalized morning dashboard showing the user’s wellbeing state, current context, and recommended 10-minute reset.

• 

  AURA compares two possible interventions using personal evidence and recommends the option with slightly stronger supporting evidence.

• 

  Guided reset session with a countdown and steps for gentle stretching, walking, and settling breathing.

• 

  Presents observed personal patterns, including relationships between workload, sleep, energy, movement, and evening recovery.

• 

  Chat interface where users can ask AURA questions about their wellbeing patterns, mornings, energy, and recent experiences.

• 

  Shows weekly wellbeing consistency, check-ins, minutes, experiments, wellbeing trends, and key discoveries.

• 

  Displays learned patterns, influencing factors, areas still being learned, strongest patterns, recent experiments, and user preferences.

• 

  Explains how AURA works, what data is stored locally, what information is collected, and the user’s option to export data.

• 

  PDF export showing the user’s profile, activity summary, wellbeing scores, personal patterns, latest check-in, and 7-day trends.

• 

  Overview of AURA’s purpose, user history, completed check-ins, experiments, engagement time, and most-used intervention.

Inspiration

Many well-being apps merely monitor separate metrics such as mood, habits, or level of activity. We therefore decided to create something more personal—a system which learns how various parts of a person's daily life affect their well-being. This led to the development of AURA, a Personal Wellbeing Twin that continuously learns from the check-ins, context, experiments, and feedback.

What it does

AURA produces a personal digital representation of the user's state of wellbeing and enables users to:

• Do daily wellbeing check-ins. Keep tabs on your mood, energy levels, sleep, stress, and daily context.
• Find your own patterns and links between actions and wellbeing.
• See a live Wellbeing Twin. Get tailored recommendations for intervention. For each recommendation, understand the reason for it. Attempt some small wellbeing experiments and give your feedback. Keep record of any progress, discoveries, and changes as they occur. Talk to AURA for personalized and context-aware guidance.

The fundamental learning loop consists of observing, understanding, identifying patterns, making a recommendation, carrying out an experiment, receiving feedback, learning and then updating the Twin.

How we built it

AURA is a full-stack web application which has been developed using React and Vite on the frontend and Node.js with Express on the backend; the frontend features authentication, onboarding, check-ins, the Wellbeing Twin, Discover, AURA Chat, Progress, and Profile screens. The backend includes APIs for users, check-ins, recommendations, experiments, patterns, progress, and AI conversations, while a structured wellbeing data model keeps record of user context, interventions, experiments, patterns, personal responses, and hypotheses. We have developed a deterministic AI component which bases its responses on the user's structured wellbeing data rather than creating new information. The prototype makes use of in-memory persistence and is intended to be expanded to include a real LLM and a database such as PostgreSQL or Supabase.

Challenges we ran into

A major challenge for us was ensuring that personalization was meaningful without making any unsubstantiated medical claims. To achieve this, AURA was designed to express uncertainty and to use phrases such as "appears to" and "moderate confidence". A further difficulty was in setting up a full Human–AI learning loop instead of just developing a chatbot; we needed the various elements such as recommendations, experiments, feedback, evidence, pattern discovery, and Twin updates to function as part of a single, continuous experience. We also needed to strike a balance between having a refined interface suitable for a hackathon and maintaining a functional full-stack architecture that could be extended beyond the prototype.

Accomplishments that we're proud of

We are proud that AURA does not merely provide wellness tracking but instead shows a true personal learning loop. Key accomplishments include:

• An interactive Wellbeing Twin experience. Making recommendations that are tailored to the user's context. — Providing explanations for the question "Why this?" Some small experiments were carried out using feedback that was Better/Same/Worse. The discovery of personal patterns and influences. Visualization of progress and trends. AURA AI Chat that is aware of the context. The full frontend and backend architecture. The design is safety-oriented and does not present wellness insights in the form of medical diagnoses.

What we learned

We realized that effective personalization involves more than just gathering greater amounts of data; the system has to understand the context, communicate the uncertainty, explain the reasons for its actions, and learn from the user's real experiences. We also realised how important it is to design AI using structured and reliable data rather than letting a language model freely produce recommendations; the complete feedback loop enabled us to see how AI can become an adaptive companion rather than merely a conversational interface.

What's next for AURA — AI-Powered Personal Wellbeing Twin

The next stage will be to turn the prototype into a platform that is ready for production. Our intention is to:

• Use a real LLM to make conversations more personal. Use Supabase or PostgreSQL instead of in-memory storage. You can connect optional wearable and activity data. Improve the Personal Utility Score and the pattern detection.
• Build better long-term wellbeing models.
• Add more small, personalized experiments. Improve the controls relating to privacy, security, and consent.
• Add support for multiple languages, including local ones. Make the uncertainty and safety measures stronger. Look into an agentic architecture which allows AURA to actively spot opportunities and lead users through personalised wellbeing experiments.

Built With

• ai
• behavoural
• chatbot
• css
• data
• design
• express.js
• full
• generative
• insights
• javascript
• mental
• node.js
• personalization
• react
• responsive
• restapi
• stack
• tailwind
• tracking
• typescript
• visualization
• vite
• wellbeing
• wellness