-
-
Rythm - Listen to the invisible.
Inspiration
We realized something strange: we track our steps, calories, sleep, heart rate, and screen time but not the organ actually running everything.
As students and professionals, we’ve all experienced the same frustration: studying for hours but retaining little, sitting in meetings while mentally drained, or pushing through work long after our brain has quietly checked out. Burnout doesn’t happen suddenly, it builds invisibly.
Rhythm was inspired by a simple question: What if your brain could tell you when to focus, when to rest, and when to stop, all in real time?
We wanted to make the invisible visible.
What it does
Rhythm is a wearable + app system that reads brainwave activity throughout the day and translates it into one usable signal: your Brain Score - a real-time indicator of cognitive readiness.
It tracks brainwave states (Delta, Theta, Alpha, Beta, High Beta, Gamma) and maps them into meaningful insights:
Detects cognitive fatigue before burnout
Identifies peak performance windows
Tracks stress escalation
Maps chronotype from real neural data
Visualizes your day as a "Brain River" timeline
Provides smart nudges based on your state
Processes everything on-device, your neural data never leaves
Instead of guessing when you’re sharp or drained, Rhythm shows you.
How we built it
We designed Rhythm as a tightly integrated hardware + software system:
Wearable EEG Interface: A lightweight headband captures electrical brainwave signals using dry electrodes optimized for everyday wear.
On-Device Signal Processing: Raw EEG signals are filtered, cleaned, and transformed using FFT and band-power extraction into recognized frequency bands.
State Classification Engine: A machine learning model translates wave patterns into cognitive states (focus, stress, recovery, flow).
Brain Score Algorithm: We compress multi-band activity into a single intuitive real-time number.
Mobile App Visualization: The Brain River timeline color-codes your day by state, revealing peaks and crashes at a glance.
Privacy Architecture: All neural data is processed locally. No third-party storage. No cloud dependency.
Our core design principle: Complex science inside. Radical simplicity outside.
Challenges we ran into
Signal Noise EEG signals are extremely sensitive. Movement, blinking, and environmental interference create artifacts. We had to build robust filtering and smoothing techniques to ensure reliable real-world readings.
Translating Data Into Meaning Raw brainwaves are not useful to everyday users. Converting complex frequency distributions into a single, trustworthy Brain Score required extensive modeling and validation.
Avoiding Anxiety Loops We discovered that constantly showing brain metrics could increase stress. We implemented an Anti-Anxiety Guard that switches to passive mode if users check too frequently.
Privacy vs. Intelligence Tradeoff Keeping everything on-device limited cloud-based modeling. We had to optimize lightweight machine learning models that run efficiently without external servers.
Designing for Trust When you’re dealing with neural data, trust is everything. We built privacy-first architecture from day one.
Accomplishments that we're proud of
Built a real-time Brain Score powered entirely on-device
Created the Brain River visualization to make neural patterns intuitive
Developed a burnout early warning system before subjective symptoms appear
Designed an ethical neural data policy: users own 100% of their data
Translated complex neuroscience into something simple enough for daily use
Most importantly, we made brainwaves actionable and no longer limited to academics.
What we learned
Burnout starts long before people feel it.
Simplicity beats feature overload.
If you show people data without context, you create anxiety — not clarity.
Privacy must be designed in, not added later.
The hardest part of building deep tech isn’t the algorithm, it’s making people trust it.
We also learned that your brain has rhythms and most people are living against them.
What's next for Rythm
Clinical validation studies for cognitive fatigue detection
Expanding stress detection with adaptive intervention protocols
Integration with calendar apps to auto-protect peak hours
Longitudinal chronotype modeling
Lightweight form-factor improvements for all-day comfort
Developer SDK for cognitive-state-aware applications
Built With
- figma
