Neuromorphic circuits are analog circuits that implement models of biological systems for sensory processing. Because the neuromorphic circuits share the same physical constraints as their biological counterparts, they have similar organizational structures, and use similar strategies for optimizing robustness to noise, and power consumption. Silicon retinas and other neuromorphic vision chips, built as hardware models of biological vision systems, represent efficient artificial sensory pre-processors. General processing networks which use detailed models of neurons are also being investigated. Simple neuromorphic systems have been built, using networks of silicon neurons and neuromorphic chips as front-ends for pre-processing incoming sensory signals. In this chapter we present two representative case studies describing a single-chip vision sensor and a multi-chip processing network, that contain most of the characteristic elements found in today’s neuromorphic systems.