Biofoundries are transforming synthetic biology, with the DNA assembly workflow being critical for successful biofoundry operation. This review examines recent advances in automated DNA assembly strategies for biofoundry, focusing on three key perspectives. First, we discuss emerging platforms ranging from high-throughput and highly efficient systems to affordable and accessible solutions. Second, we explore how standardized design tools enable seamless interoperability across diverse biofoundries, facilitating protocol sharing and reproducibility. Third, we analyze the integration of machine learning into assembly workflows, in which AI-driven systems dynamically optimize protocols, diagnose failures, and close the DBTL loop through real-time learning. These convergent advances are establishing a new paradigm in which experiments continuously improve through iteration, promising to accelerate both fundamental research and industrial applications.