Deep learning has changed the landscape for natural language processing (NLP) technologies, which enable quickly extracting valuable insight from enormous volumes of text. The tremendous potential of deep learning has become real in just the last seven to eight years, but what does it take to bring this technology out of the lab and into the field? This talk will discuss the work up until now in deep learning that has focused on improving accuracy. These leaps in accuracy have everyone eager to get deep learning deployed in the field, but there are still several hurdles. I will talk about the issues that need to be surmounted for deep learning to become feasible and practical. We will survey the research work of Basis Technology and others to boost deep learning’s capabilities in speed, scalability, and most importantly, language coverage.