Large-scale brain mapping via brain-machine interface is important for deciphering neuron population dynamics, understanding and alleviating neurological disorders, and building advanced neuroprosthetics. Ultimately, brain mapping aims to simultaneously record activities from millions, if not billions, of neurons with single-cell resolution, millisecond temporal resolution and cell-type specificity over the time course of brain development, learning, and aging. In this talk, I will first introduce “tissue-like” soft bioelectronics that possess tissue-like properties, capable of tracking the electrical activities from the same neurons in the brain of behaving animals. Specifically, I will discuss the fundamental limits to the electrochemical impedance stability of soft electronic materials in bioelectronics and introduce our strategies to overcome these limits, enabling a scalable platform for the large-scale brain mapping. Then, I will discuss the building of “cyborg organisms”, where stretchable mesh-like electrode arrays are embedded in 2D sheets of stem/progenitor cells and reconfigured through 2D-to-3D organogenesis, enabling continuous 3D brain electrophysiology during brain development. Finally, I will discuss future perspectives that leverage the soft bioelectronics-brain interface to integrate single-cell spatial transcriptomics with electrical recording, opening opportunities for cell-type-specific brain mapping and functional brain cell atlas