Edge AI and On-Device Intelligence

A premium, hands-on course on Edge AI and on-device intelligence, updated for 2026. You will learn why intelligence is moving off the cloud and onto phones, microcontrollers, wearables and industrial gateways — driven by latency, cost, privacy and offline reliability — and how to actually ship it. You will master model compression (post-training and quantization-aware quantization, pruning and sparsity, knowledge distillation), the real runtimes (LiteRT / TensorFlow Lite, ONNX Runtime, Core ML, ExecuTorch, MediaPipe and llama.cpp), and mobile deployment on both Android and iOS. You will go all the way down to TinyML on microcontrollers, understand the edge hardware landscape (NPUs, Google Coral, NVIDIA Jetson), and run small language models such as Gemma, Phi and small Llama models directly on device. The course covers on-device inference optimization, privacy-first design and federated learning, over-the-air model updates, and edge monitoring, with real case studies across mobile, IoT and wearables. Every concept is paired with real, correct code, and privacy-by-design and GDPR obligations are treated as first-class engineering concerns. Includes a comprehensive final assessment.