Inspiration
Caught up in our daily commitments, we rarely find time to reflect on the stressful moments that accumulate throughout the day, leaving feelings of overwhelm that are difficult to shake. Journaling has proven to be an effective way to decompress, so we wanted to explore ways to enhance its therapeutic impact. While many apps focus on helping users calm down after stress, we believe it is far more important to prioritize prevention rather than just treatment when it comes to mental health.
What it does
kAIro is an intelligent journaling and productivity app that synchronizes a student’s mental health with their cognitive performance. It will track emotions through journaling, with an adaptive background that visualizes their mood and provides background music, making the experience much more immersive. It dynamically schedules tasks based on cognitive load and energy levels, helping students stay focused and mentally resilient.
How we built it
We built Kairo, a productivity and wellness app designed to help students balance focus and mental health. Developed with a modern full-stack setup using React, TypeScript, and Vite, most of which we vibecoded in Cursor, Kairo features a polished interface built with Tailwind CSS and shadcn/ui. It runs on Lovable Cloud (Supabase), providing PostgreSQL storage, serverless edge functions, and secure RLS-based authentication. Through the Lovable AI Gateway, we were able to access Google Gemini and OpenAI models without managing API keys. We created Deno-based edge functions like extract-tasks for mood and task detection, optimize-tasks for prioritization, scan-schedule for vision-based event extraction, and process-transcript for turning ElevenLabs voice chats into structured tasks. Kairo’s journal analyzes emotions in real time, using animated gradients and mood chips that respond as users type, while its dashboard tracks stress through incomplete tasks and recent moods. The ElevenLabs Conversational AI assistant securely converts voice interactions into actionable items, and with React Query and Supabase subscriptions, Kairo delivers real-time updates. We also integrated CSV task imports, OCR-based schedule scanning, and mood-based Spotify playlists, all crafted to make productivity feel intuitive, emotionally aware, and deeply human.
Challenges we ran into
During the process of developing kAIro, we faced two major challenges. The first one was narrowing down our product features and making sure the app stayed simple, yet effective. While there are a lot of variables that go into mental health struggles, we decided to aid in task prioritization and provide our users with an immersive and safe atmosphere in which they are able to share their struggles and get real-time feedback. The second challenge was the process of building kAIro itself. Since many of us used this opportunity as our first vibecoding experience, it took us time to optimize our prompt engineering processes to get our desired look for kAIro on the screen.
Accomplishments that we're proud of
We’re proud that kAIro began as a personal struggle. Each of us has felt the burnout and emotional fog, and we are happy to have turned that shared experience into something meaningful. We are extremely proud that we were able to build a working prototype that is able to listen and respond to emotion.
What we learned
The most important takeaway was that innovation is not just about creating new technology. It is about continuously trying to improve and integrate it to elevate the human experience. We grew together as a team, learning to communicate and support each other through the late-night sprints.
What's next for kAIro
Looking ahead, we want to make kAIro feel even more personal and effortless. By integrating speech-to-text recognition, users will no longer need to type out their tasks. They can simply speak, and kAIro will take care of the rest. We also hope to make kAIro a space that celebrates positivity, helping users capture moments of joy and gratitude so they can return to them when life feels overwhelming.
Built With
- 11labsapi
- cursor
- geminiapi
- javascript
- lovable-backend
Log in or sign up for Devpost to join the conversation.