Videquus is a horse tech start-up that aims to help users take better care of their horses using camera surveillance. Piktiv was hired to quickly build the first version of the client's product.
Piktiv was responsible for all software development in the first production of the software. The product is delivered using a custom-built camera unit that is shipped to the client as part of a subscription agreement and connects to an app. With the app, users can watch their horses live, receive a well-being report, and send notifications for special events during the night.
Piktiv developed the deep learning vision algorithm that uses image segmentation to identify the horse and its actions. We built all the necessary infrastructure to provide this service to the end-user, including software running on the edge surface, a backend system to the cloud, and the mobile app. The deep learning model is continuously trained using the supportive infrastructure. The key reason why we could do this so well is that we have broad knowledge in working with different technologies, minimizing misunderstandings, and ensuring we could continuously deliver vertical slices to the client.
The result was a ready-to-deliver system executed according to the roadmap. The short feedback cycles during the process allowed the product to evolve according to customer needs.