the Noise
OVERVIEW.
Twylit is an AI soundscape engine that personalizes sleep.
Millions of people struggle with poor sleep, often waking up tired and unrefreshed despite trying countless solutions. Existing tools are one-size-fits-all and fail to adapt to individual sleep patterns and preferences, leaving users without truly effective support.
Twylit solves this challenge by generating adaptive, AI-driven soundscapes tailored to each user’s unique sleep profile, promoting deeper, more restorative rest.
During a productathon, I collaborated with a cross-functional team to ideate and prototype the platform, leading the design of the personalized soundscape experience.
By personalizing sleep routines, Twylit helps users fall asleep faster and wake up more refreshed, creating a scalable solution for better sleep health.
PROBLEM.
Many individuals struggle to get a restful night’s sleep, often waking up feeling tired and unrefreshed. Existing sleep tools are rigid and impersonal, offering one-size-fits-all solutions that don’t adapt to individual patterns or preferences.
Users needed a system that could intelligently enhance their sleep while still allowing control over the sounds and environment that help them relax.
RESEARCH.
But what exactly was keeping people up at night?
We began with a clear understanding that most users struggled to achieve consistent, high-quality sleep. To discover why, we combined direct user feedback with in-depth exploration of the science behind sleep and sound.
After gaining insights from 34 participants, we discovered that:
- Temperature and noise were top sleep disruptors
- White noise improved sleep quality
- Average restful nights: 3.44 per week
Our external research further revealed that different noise patterns have distinct effects on sleep. Some sounds help users fall asleep faster, while others help maintain deep, uninterrupted rest.
These findings highlighted the need for a personalized, adaptive approach to improving sleep.
INITIAL DESIGNS.
Creating a personalized sleep application that is modern, intuitive, and easy to use. I designed the initial screens to be minimalistic, with a focus on the main features and the user's sleep patterns.
SLEEP MARKET RESEARCH.
With over $18 billion in the sleep tech market, but no solution that aims to enhance sleep, we knew we had a unique opportunity to create a product that would be both effective and popular.
The sleep tech market is projected to surpass $300 billion by 2035, showing no signs of slowing down.
Sound machines currently make up only 8% of the sleep tech market, leaving significant room for innovation and growth in this space.
By entering this market, our product is set to capture a share worth up to $14 million, with clear potential to scale alongside industry demand.
FINAL DESIGNS.
Our research-driven approach resulted in a modern, intuitive interface that feels both sleek and approachable. The final Twylit design combines clean visuals with thoughtful interactions, making it easy for users to customize their sleep environment while benefiting from AI-driven personalization. By blending aesthetics with functionality, the platform delivers a seamless, restorative experience that adapts to individual sleep patterns and preferences.
WHAT I LEARNED.
Working on Twylit was an intense, hands-on lesson in turning insights into creative, technology-driven solutions. The 48-hour hackathon pushed us to iterate quickly, test ideas in real time, and gather feedback efficiently, teaching me how to make thoughtful design decisions under pressure.
By diving into user research and studying sleep patterns, I was able to design an AI-powered system that adapts to individual needs while giving users control over their sleep environment. Balancing technical possibilities, user experience, and real-world constraints pushed me to think holistically; considering not just the interface, but the full workflow and how each feature impacts engagement.
The experience taught me the value of rapid prototyping and the importance of gathering user feedback early and often. It also reinforced the importance of considering not just the interface, but the full workflow and how each feature impacts engagement.