Inspiration
Sonder started with my grandfather. He lived with Alzheimer disease and we watched his mind change slowly in ways we did not understand at the time. Seeing how those small shifts in memory and attention slipped by unnoticed made me want to build something that pays attention earlier and gives families more time to notice change.
What it does
Sonder learns your personal baseline using data from your eyes, voice, cognition, and health signals and then tracks how far today drifts from that baseline. It turns short interactive scans and your daily patterns into one brain drift score plus a simple explanation and one helpful action that protects your focus or recovery.
How we built it
We built Sonder as a native iOS experience that feels calm and premium with soft motion curved surfaces and a simple orb based entrance instead of a clinical form. Underneath that we spent a lot of time stitching together camera based eye tracking voice capture cognitive tasks and health data into one consistent timeline so the app can quietly learn how your brain usually functions.
Challenges we ran into
The hardest part was staying honest about what the data can and cannot say so Sonder shows trends and estimates without ever sounding like a diagnosis. We also had to design the whole flow so that permissions health data and scans work together smoothly while still keeping the app fast private and gentle enough that people would actually use it every day.
Accomplishments that we are proud of
Sonder gives a clear brain drift score and real evidence with almost no manual logging or journaling from the user. We are especially proud that the experience feels like a quiet operating system for your mind where you can see your eyes being tracked your focus and reaction changing over time and one small action that actually fits your own patterns.
What we learned
We learned that early cognitive changes are subtle noisy and heavily influenced by sleep stress and lifestyle so any honest app must speak in gentle signals not hard labels. We also learned how much thought it takes to turn raw signals into something that feels human understandable and kind instead of overwhelming or scary.
What is next for Sonder
Next we want Sonder to learn which specific actions truly reduce drift for each person and to make those recovery suggestions smarter and more personal over time. We also plan to deepen how we use wellbeing data and to work with researchers so that one day our drift metrics can be carefully compared with formal cognitive tests and real world outcomes.
Log in or sign up for Devpost to join the conversation.