To elucidate the relationship between microstructural characteristics and mechanical properties in additively manufactured (AM) maraging steel, this study introduces a computational approach that addresses two fundamental challenges. Firstly, it addresses the creation of representative volume elements (RVEs) that mimic the observed microstructural complexities, such as meltpool boundaries, prior austenite grains, packets and blocks of lath martensite. This is accomplished through the application of Potts Monte-Carlo methods and grain segmentation techniques in accordance with the Kurdjumov–Sachs orientation relationship. Secondly, this study develops a comprehensive crystal plasticity (CP) model encompassing both bcc and fcc plasticity. Inspired by atomistic and discrete dislocation dynamics studies, the proposed CP model incorporates characteristics intrinsic to bcc plasticity, including non-Schmid effects, dislocation and precipitate strengthening, and Hall–Petch type strengthening of elongated martensitic blocks. Utilizing the created RVEs and the proposed CP framework, finite element simulations are conducted based on an update-Lagrangian formulation. The purpose of this study is to investigate the deformation behavior, texture evolution, tension–compression asymmetry, and evolution in dislocation density in RVEs representative of as-built and heat-treated samples of maraging steel. This computational approach and its findings deepen our understanding of the intricate interplay between microstructural characteristics and mechanical properties in maraging steel and also provide valuable guidelines for refining its additive manufacturing and heat treatment processes.