In this article, we demonstrate how to implement a version of deep Q-learning, a reinforcement learning technique, to create an AI agent that can play checkers at a reasonable level.