Optical imaging systems have evolved with the goal of producing an isomorphic measurement of a scene. Typically such imaging systems place the sole burden of image formation on optics while the role of detector array is relegated to sampling/digitization of the optical image. Post-processing is usually viewed as a “tool” to mitigate image artifacts/noise, apply compression, and/or enable exploitation tasks such as pattern recognition, target tracking etc. The traditional design approach optimizes each sub-system (optics, detector, post-processing) separately and often results in sub-optimal designs. In contrast, computational optical imaging exploits the optical, detector, and post-processing design degrees of freedom jointly to achieve end-to-end system optimality. Such a joint design approach is especially suited to task-specific imaging as it allows one to incorporate knowledge of scene statistics and specific task in the system design. In this talk, I will discuss examples of computational imaging system design for specific tasks, such as image formation and pattern recognition, to highlight the power of the joint-design framework. A task-specific information-theoretic approach to imaging system design and analysis will be also discussed in the context of the fundamental limits of imaging systems.