Researchers have developed a framework called TO-Agents that connects human design intent with topology optimization. The system uses natural-language input to guide the optimization process and includes multiple agents that critique and revise solver parameters. In two case studies, the system successfully produced preference-aligned designs in 60% of trials, outperforming an ablated pipeline without visual or historical feedback. TO-Agents also enables end-to-end intent-to-prototype design by incorporating a manufacturing agent. --- Why it matters: This matters to engineers and researchers because it has the potential to shift the focus from low-level parameter tuning to higher-level specification of form and function, making design more efficient and effective. Source: https://arxiv.org/abs/2605.21622