Urban design is reimagined as a parametric and generative process where spatial form emerges from the dynamic interplay of data, constraints, and simulation. At the core of this approach is the use of tensor fields—abstract vectorial representations of forces such as terrain, access, density, and amenity proximity—that enable flexible, context-sensitive modeling of cities. Rather than drawing static masterplans, designers define spatial intentions through gradients and inputs, which are then computationally interpreted to produce street networks, blocks, and buildings. This method allows for rapid generation of diverse urban configurations, all subject to performance metrics such as walkability, energy demand, solar access, and spatial accessibility. Design is not linear but iterative and exploratory, framed as a search for optimal trade-offs within a complex, multi-objective space. Urban form is treated as a dynamic system, where small parametric changes can lead to wide morphological variation and emergent spatial qualities. By encoding zoning, environmental factors, and mobility goals into tensor maps, the system democratizes urban modeling, allowing designers to engage with wicked problems through simulation rather than intuition alone. The emphasis shifts from form-making to system-thinking, from solution to possibility space, fostering a richer, data-informed dialogue between design agency and environmental constraints.
