Edge AI
Running AI models directly on local devices (phones, IoT sensors, cameras) rather than sending data to the cloud for processing.
Why Edge?
Lower latency (no network round-trip), works offline, preserves data privacy (data never leaves the device), reduces cloud costs, and enables real-time applications like autonomous driving and industrial robotics.
Challenges
Edge devices have limited compute, memory, and power. Models must be optimized through quantization, pruning, distillation, and architecture-specific compilation (ONNX, TensorRT, Core ML).
Examples
Face ID on iPhones (on-device face recognition), smart cameras with person detection, voice assistants processing wake words locally, predictive maintenance sensors in factories, and autonomous vehicle perception systems.