WiFi-based human pose estimation offers a privacy-preserving and cost-effective solution for human motion monitoring, with significant implications for enhancing energy efficiency in smart buildings. In the context of smart energy systems, the use of low-power WiFi devices is crucial for sustainable energy management. However, deploying WiFi-based human pose estimation on these resource-constrained devices presents challenges due to computational limitations. To address this, we propose EfficientWiPose, a model specifically designed for commercial low-power WiFi devices. EfficientWiPose achieves low model complexity and hardware-friendliness, facilitating the integration of human motion monitoring into smart energy systems. Our experiments show EfficientWiPose achieves an impressive mPCK@0.05 of 78.67%, outperforming other WiFi-based methods. Its hardware-friendly and energy-efficient design provides a robust solution for accurate pose estimation and paves the way for real-time WiFi sensing in various practical applications, especially in smart buildings where reducing energy consumption is paramount.