Kinematic recommendation across devices using machine learning

A computer-readable, non-transitory storage medium storing computer-executable instructions, which when executed cause one or more processors to perform:

• receiving a data point generated by one or more sensors, including subject data related to a subject and field data related to a field in real time;
• generating, based on the data point, a current state;
• identifying a kinematic recommendation based on the current state using a first machine learning model;
• determining a selection of one or more output devices of a plurality of output devices of a plurality of types for communicating one or more portions of the kinematic recommendation and a timing of communication for each selected output device, based on the current state, the kinematic recommendation, and characteristics of the plurality of output devices using a second machine learning model;
• transmitting the one or more portions of the kinematic recommendation respectively to the one or more output devices in accordance with the timing.