The goal of EM neural reconstruction, or connectomics, is to discover the neural wiring diagram from Electron Microscopic (EM) images of animal brain tissue. Computational framework for EM reconstruction typically consists of machine vision tasks such as volume segmentation, synaptic connectivity detection etc. In this talk, I will present the insights gathered from multiple connectomic reconstructions from EM images collected from different animals with different imaging techniques. In particular, I will emphasize on defining target functions (analytical or non-analytical mapping) for learning segmentation, synapse detection to achieve optimal performance. The presentation will include quantitative and qualitative results from methods designed for different connectomic tasks as well as some discussions on large scale computation and potential future directions.