EpiSafe

Epilepsy is unpredictable by definition. For millions of people, the hardest part isn't the seizure itself — it's the hours, days and nights spent not knowing when the next one might come, and whether anyone will be there if it does.

Existing monitoring is either clinical and bulky, or consumer-grade and unreliable. There was a clear gap: something wearable, accurate, and genuinely trustworthy — designed around the person wearing it, not the lab that built it.

Before drawing a single part, I ran 40+ structured interviews with patients, caregivers and clinicians. The goal wasn't to validate an idea — it was to understand the real texture of living with seizure risk.

Those conversations reshaped the product more than once. Comfort, discretion and battery anxiety turned out to matter as much as raw sensor accuracy. The brief came from people, not assumptions.

Three prototype generations, each built to answer a specific question: can it sense reliably, can it last a day, can someone forget they're wearing it?

Each iteration was rough on purpose — built to be proven wrong quickly, then refined. Below: the hardware as it evolved.

EpiSafe pairs continuous biosignal sensing with on-device detection and an emergency-alert path to a trusted contact. The engineering challenge is the balance: sensitive enough to catch events, specific enough to earn trust by not crying wolf.

Alongside the hardware, I've been doing the early FDA 510(k) pathway groundwork and risk analysis — because in medical devices, the regulatory story is part of the product.

Next: a validation study with real users, tighter detection models, and a manufacturable industrial design. The north star hasn't changed — make something people can genuinely rely on.