Deep neural learning (DNL) is a growing algorithm used in various different application areas (i.e. natural language processing, computer vision, image processing, cybersecurity, handwriting recognition, speech recognition etc.). However, field devices (i.e. sensors, cameras etc.) capture real time data which is analyzed by the deep learning models usually remotely. Although input-data must be processed and analyzed in run time but all time it is not possible because servers are placed remotely and interruptions are possible during data transmission (i.e. slow network, denial of service, electricity, bandwidth issue etc.). Thus, the placement of computing devices near to field devices or locally analysis of the captured data can help in this situation. Here the edge computing comes in picture, which allow the placement of deep learning systems near the field device and allow to take decisions locally with the analysis of locally captured data which enhance the latency and competence of the work and evaluation making. This chapter is exploring the amalgamation of edge computing with deep learning framework and their benefits.