Rafael Pastrana
︎ arpastrana@princeton.edu
︎ @arpastrana
︎ @arpastrana
︎ Publications
Rafael is a Ph.D. candidate at Princeton. He develops auto-differentiable tools for structural design, like JAX FDM and COMPAS CEM.
Rafael currently works as a software engineering intern in computational geometry at Robert McNeel & Associates.
Rafael holds a Master of Advanced Studies in Architecture and Digital Fabrication from the Swiss Federal Institute of Technology (ETH) Zurich and a Master of Arts from Princeton.
For over 3 years, he worked at Bollinger + Grohmann, where he was responsible for the structural analysis and geometric rationalization of art installations in Europe and in Australia.
Rafael worked at the Block Research Group. There, he contributed to the development of a COMPAS package to streamline the generatation of the fabrication data for one of the full-scale prototypes of the functionally-integrated, funicular floors at the NEST HiLo.
At Gramazio Kohler Research, and in collaboration with Autodesk, Rafael trained a reinforcement learning model that overcame geometrical and material inaccuracies arising during the robotic assembly of timber structures.
︎ @arpastrana
︎ @arpastrana
︎ Publications
Rafael is a Ph.D. candidate at Princeton. He develops auto-differentiable tools for structural design, like JAX FDM and COMPAS CEM.
Rafael currently works as a software engineering intern in computational geometry at Robert McNeel & Associates.
Rafael holds a Master of Advanced Studies in Architecture and Digital Fabrication from the Swiss Federal Institute of Technology (ETH) Zurich and a Master of Arts from Princeton.
For over 3 years, he worked at Bollinger + Grohmann, where he was responsible for the structural analysis and geometric rationalization of art installations in Europe and in Australia.
Rafael worked at the Block Research Group. There, he contributed to the development of a COMPAS package to streamline the generatation of the fabrication data for one of the full-scale prototypes of the functionally-integrated, funicular floors at the NEST HiLo.
At Gramazio Kohler Research, and in collaboration with Autodesk, Rafael trained a reinforcement learning model that overcame geometrical and material inaccuracies arising during the robotic assembly of timber structures.