This dissertation presents a new method called “Autonomous Engineering” that incorporates geographic information systems (GIS) to automatically design green stormwater infrastructure. The Autonomous Engineering framework aims to increase the efficiency at which green infrastructure is designed, thus promoting increased implementation. Green infrastructure design is a unique challenge in that it is multi-scale; planning and design considerations must be made at both the site-level and the watershed level by analyzing various types of spatial data. This framework presents a methodology for designing green infrastructure based on a combination of remotely sensed watershed-scale data and ultra-high resolution site-level Light Detection and Ranging (LiDAR) data. First, watershed level data is analyzed to generate site recommendations and quantify runoff characteristics. Second, LiDAR data is processed using both deep learning and machine learning frameworks so that site-level spatial features can automatically be recognized and extracted and so that an ultra high resolution digital elevation model (DEM) is generated. Next, linear referencing techniques are used to analyze terrain and identify geometric design recommendations. The results are finalized in the form of custom design drawings and reports. This work has outcomes for improved green infrastructure design workflows as well as the spatial analysis of robust site-level data for other applications. Future work includes the extension of these methodologies to applications beyond green infrastructure.