Dogs are incredible companions, but as many pet owners know, they can be remarkably skilled at hiding when something is wrong. My own Labrador-Great Dane mix, for instance, contracted tick fever several times over his 14 years, and it was always at least a few weeks before I noticed anything off. It was a source of constant anxiety.
Hoomanely, a new startup based in Palo Alto, aims to help owners like me catch health issues in their dogs much earlier. The company has developed an AI-powered platform that collects data through a sensor-laden feeding station called the EverBowl. This station measures food and water intake, eating speed, and records chewing and swallowing sounds, facial thermal patterns, oral movements, and other signals.
Once the data is collected, Hoomanely's AI analyzes it to establish a baseline for each dog and builds a comprehensive health record. Any significant or prolonged deviations from this baseline are flagged to the owner via a mobile app and in long-term reports. The company says these reports can provide valuable data to veterinarians when needed. The app also allows users to input additional information, ensuring the baseline stays accurate even for signs outside the sensors' scope.
“Practically, the bowl is the best capture environment in a pet’s life: same place, same posture, same routine, twice a day, for years,” said Sai Supriya Sharath, co-founder and CEO of Hoomanely, in a recent interview. She emphasized that appetite, hydration, and oral comfort are often among the first things affected by pain, dental disease, digestive issues, endocrine changes, or illness. “Dogs are evolved to mask lethargy and limping. They are far less able to mask how they eat and drink… reduced or altered intake and changed drinking are presenting signs across a wide range of conditions.”
Hoomanely’s platform was developed after an 18-month beta testing phase, during which the company collected around 5 million data points from over 80 dogs. During this testing, the system detected changes in one dog's eating patterns that were later linked to an infected chipped tooth. In another case, the app showed persistent changes in a dog's eating and temperature baselines, leading the owner to seek veterinary care and receive a diagnosis of tick fever.
While the potential is clear, there are some limitations. The beta cohort was relatively small, and the company has only shipped about 50 devices so far. Currently, the EverBowl and app are available through a $29-per-month subscription. Sharath also noted that Hoomanely has not yet published independent sensitivity, specificity, or false-positive rates for detecting clinical events. That would require a formal study comparing the system's alerts to actual diagnoses. The company acknowledges that its platform is not a replacement for professional veterinary diagnosis, but rather a tool to help owners and vets identify issues more quickly.
Looking ahead, Hoomanely is using its data to conduct formal veterinary studies to validate the system's effectiveness across a larger, more diverse group of dogs. “The veterinary studies being designed will measure sensitivity, specificity, positive predictive value, and false positives for each alert category, and we intend to report results by use case rather than as one headline number,” Sharath said.
As the pet tech industry continues to grow, innovations like the EverBowl represent a promising step toward proactive pet healthcare. With more rigorous testing and validation, such AI-driven tools could become essential for early disease detection, giving owners and veterinarians a better chance at timely intervention.