Position control of a brushless DC motor is challenging as it requires a high degree of accuracy with nearly zero steady-state error and no overshoot. This paper presents data-driven control (DDC) of a brushless DC motor using neuro fuzzy proportional-integral (PI) controller. To eliminate the difficult issues of model-based control, the controller is designed by using an input–output data set obtained from a controlled system. The data set for the proposed autotuning method is obtained from the conventional PI controller tuned under changing load torque conditions. Every time the load torque varies, retuning of the PI controller is required, which is a tedious task. The neuro-fuzzy system helps in auto tuning the PI controller. The proposed methodology is implemented using MatLab Simulink. Simulation results show an improvement in rise time, settling time and steady-state error with the proposed controller in comparison to the conventional PI controller.