Now that you have an AI Engine application, the next step is to find a way to send the input data and store the output data on hardware. This is accomplished by creating PL master kernels that store the input data in block RAM and feed it to the AI Engine. Reference inputs are used in this design to mimic data flow that would come from a radio antenna. PL slave kernels, which receive AI Engine output data and store it into URAM resources, must also be created.