Pragmatic approaches to solving problems in artificial intelligence arise due to fast technological evolution as well as the complexity of real-world tasks. These operational, bottom-up approaches are based on representing how to behave in specific contexts. In this chapter, pragmatic approaches are discussed from the perspective of agents (human and AI) using data, information, and knowledge in context. Information is what is interpreted by an agent from data about the world, and what is exchanged with other agents. Information is highly context-dependent and under the control of the knowledge that is in the agent's "mind". Context plays a pivotal role in the development of information and knowledge. Setting data, information and knowledge in context allows new approaches to modeling (human and AI) reasoning. Based on its contextual knowledge (the part of its knowledge concerning the current focus), a reasoner interprets and uses information and produces new information and knowledge. Concretely, the pragmatic approach is shown in three examples of context-based research addressing applications in the real world. Operational knowledge in these approaches leads to contextualization of reasoning and opens the door to context-based AI systems.