Posts by dev@insidea.com

In factories, warehouses, construction sites, and field operations, AI is moving from the cloud to the edge. 

AI solutions such as visual inspection, predictive maintenance, and real-time diagnostics are happening on the factory floor and in the field now, not by connecting to an LLM in the cloud but by running AI Agents locally on rugged tablets. 

If you’re comparing a rugged tablet against a mini PC with a webcam, or considering deploying a consumer device, this should change your mind.

Explaining “TOPS”

If you’ve been shopping for a rugged tablet or an edge AI device, you’ve definitely seen terms like “12 TOPS” or “15 TOPS” referenced. TOPS stands for Tera Operations Per Second, which means how many trillion (yes, you read it correctly) calculations an AI chip can process every second. A higher TOPS rating means the device can process AI faster.  But high TOPS devices are very expensive and rarely needed for typical Edge AI applications. 

Note: TOPS is only relevant when you run AI on the device – what we call Edge AI – not when connecting to AI in the cloud, where the computation occurs on remote servers in data centers.