Gadgetory

in a highly controlled environment like a construction site being able to track equipment people and what they're doing it's critical to workplace safety using our technology right now both people and object recognition models are currently deployed in that construction yard and then on the screen there before you you can see the solution is recognizing Jana in real time take note again of that jackhammer in a uba's workshop it is resting vertically against the bench which is actually a very unsafe position for such a large heavy object to be in and so I've already gone ahead and I've tagged that jackhammer in a variety of safe and unsafe positions in the system and so when I update this pipeline which I'm going to do right now it's not only going to see the jackhammer now but it is also going to tell me it's an unstable orientation and it's going to notify yuba so he's able to very quickly resolve the issue the solution is running more than 27 million recognitions per second across people objects and activities both in the yard and in the workshop and we're making changes to the solution in real time in the cloud and deploying it to the edge which is a great example of how we can stretch the cloud to create a mesh of interconnected devices and services