This paper describes the design of a multivariable, constrained Model Predictive Control (MPC) system for torque tracking in turbocharged gasoline engines scheduled for production by General Motors starting in calendar year 2018. The control system has been conceived and co-developed by General Motors and ODYS. The control approach consists of a set of linear MPC controllers scheduled in real time based on engine operating conditions. For each MPC controller, a linear model is obtained by system identification with data collected from engines. The control system coordinates throttle, wastegate, intake and exhaust cams in real time to track a desired engine torque profile, based on measurements and estimates of engine torque and intake manifold pressure. The MPC optimizes torque tracking during both transient and steady-state operations, minimizing specific fuel consumption and taking into account predefined fuel-efficient steady-state actuators positions, as well as constraints on input and output variables. Actuator commands are computed by solving an optimization problem at each sampling instant. Each linear MPC controller is equipped with a Kalman filter to reconstruct the system state from available measurements. Compared to more classical controls, the presented MPC approach achieves better coordination of multiple actuators for improved fuel economy and drivability, while maintaining robustness with respect to measurement noise, ambient conditions, and part-to-part variations. Moreover, the systematic, model-based framework developed for production enables an immediate adaptation of the design to different engine hardware architectures. ODYS developed the MPC core software that could be run in an embedded controller and was configurable for different problem formulations. General Motors and ODYS worked together to integrate the MPC core software to meet the requirements of various projects, improve the run-time performance and to deploy and validate in production engine control units (ECUs).