On-Device AI / Edge AI
Also written as On-Device AI, Edge AI, TinyML, Core ML, TensorFlow Lite, LiteRT, ONNX, ONNX Runtime
Running AI models directly on a phone, laptop, camera or car rather than in the cloud. Responses are faster, it works offline and data stays private, but the model must be shrunk to fit tight memory and battery limits. Core ML (Apple), LiteRT (formerly TensorFlow Lite) and ONNX Runtime are common tools.
Think of it like
Carrying a pocket dictionary instead of phoning a library every time you need a word.
Junior or senior?
Sits between ML and mobile or embedded engineering, and strong candidates have a foot in both.
Senior sounds like
Can give numbers for model size, speed and battery cost on the target device, and describe what accuracy they gave up to make it fit.
Ask them
“What device did it run on, how big was the model, and what did you trade away to make it fit?”