Modern Machine Learning (ML) approaches like Deep Neural Networks (DNNs) have shown tremendous improvement over the past years to achieve a significantly high accuracy for a certain set of tasks, like image classification, object detection, natural language processing, and medical data analytics. However, these DNN require huge processing, memory, and energy costs, besides being vulnerable to several security threats. This talk will present challenges and cross-layer frameworks for building highly energy-efficient and robust machine learning systems for the tinyML and EdgeAI applications, which jointly leverage optimizations at different software and hardware layers, e.g., neural accelerator, memory access optimizations, approximations, hardware-aware NAS and network compression. These cross-layer techniques enable new opportunities for improving the area, power/energy, and performance efficiency of systems by orders of magnitude, which is a crucial step towards enabling the wide-scale deployment of resource-constrained embedded AI systems like UAVs, autonomous vehicles, Robotics, IoT-Healthcare / Wearables, Industrial-IoT, etc.