Luminary published a blog unpacking two ways models gain spatial context in physics AI models, including gathering neighboring points across multiple scales and encoding position as a spectrum of frequencies.

Luminary enables engineers to encode physics into AI, transforming how physical systems are designed, built, and operated. What once required thousands of simulations and months of iteration can now be explored in seconds, unlocking a new era of speed, scale, and ambition in engineering. Physics AI models enable designers and engineers to instantly predict the physical performance of products like cars, aircraft, electronics and turbomachinery. Customers span industries from aerospace and defense, automotive, industrial manufacturing and sporting goods, including Otto Aerospace, Joby Aviation and Sceye. We are hiring! Check out our open roles here: https://ats.rippling.com/luminarycloud/jobs