We present the result of research conducted at deepsense.ai, that focuses on distributing a reinforcement learning algorithm to train on a large CPU cluster