Building Pet Health Tech: ML, Sensors, and Dog Behavior Data
In this podcast episode, we’ll be joined by Sofya Yulpatova, Founder and CEO of a PetTech startup building what many describe as an early version of the “Apple Watch for dogs.” Her work sits at the intersection of machine learning, sensor data, and real-world behaviour patterns, and she brings a refreshingly honest view of what it takes to make pet health measurable.
We'll discuss the challenges in animal health technology compared to human wearables, such as dogs' unpredictable behavior and the difficulty of collecting useful data. Sofya will explain why many early health signals are often undetectable by owners but clear in the data.
We’ll also cover the technical side, including developing models for different dog breeds, managing sensor noise, and creating feedback loops with veterinarians and pet owners.
Topics we plan to explore:
Why sleep patterns may be the strongest and most overlooked health indicator
How small daily behaviour changes can reveal early discomfort
The realities of collecting embedded sensor data in the real world
Challenges around calorie estimation, device variability, and multi-dog generalization
Lessons from early prototypes and testing with real pets
This episode explores how machine learning, sensor data, and behavioral science intersect, demonstrating how applied machine learning can advance pet health technology to meet expectations similar to those of human wearables.
About the Speaker:
Sofya Yulpatova is the Founder and CEO of Fit Tails, a PetTech startup developing an activity and health tracker for pets. She has a background in computer science, machine learning, and product management, and previously managed product and delivery operations at FixParts, an international automotive parts distributor. Sofya studied at the University of Latvia and completed the Sales and Marketing Programme at the Stockholm School of Economics in Riga.
DataTalks.Club is the place to talk about data. Join our Slack community!