Manipulating quantum many-body states is a central milestone en route to harnessing quantum technologies. However, the exponential growth of the Hilbert space dimension with the number of qubits makes it challenging to classically simulate quantum many-body systems and consequently, to devise reliable and robust optimal control protocols. I will present a novel framework for efficiently controlling quantum many-body systems based on deep reinforcement learning (RL). Applications include the design of entangling two-body gates which outperform state-of-the-art pulse sequences used in superconducting qubits platforms, and the construction of circuit-based protocols to prepare ground states of quantum spin chains away from the adiabatic regime using ideas from counter-diabatic driving. To tackle the quantum many-body control problem, we leverage matrix product states (i) for representing the many-body state and, (ii) as part of the trainable machine learning architecture for our RL agent. In particular, we demonstrate that RL agents are capable of finding universal controls, of learning how to optimally steer previously unseen many-body states, and of adapting control protocols on-the-fly when the quantum dynamics is subject to stochastic perturbations.