Moore’s law is approaching the end. The computing power could no longer rely on ever increasing speed. The computing architecture needs to be changed to agile, domain specific architecture. Aupera has innovated a whole new distributed micro-node architecture based on Xilinx Zynq® UltraScale+™ MPSoCs and leverage Xilinx ML tools to build an ultra-high density computing platform for highly efficient video transcoding & real-time analytics. This talk will focus on explaining details of how this new architecture best utilize Xilinx MPSoC features to change the video AI computing landscape.