Summary
Pedro Brandimarte is a research engineer with 13 years of experience blending computational physics, mathematical modeling, and high-performance software engineering. He has moved from theoretical multiscale simulations for nanotechnology to practical acceleration of deep learning on accelerators, most recently refactoring and optimizing tiling, IR parsing, scheduling and benchmarks for MindSpore at Huawei’s Paris Research Center. Pedro pairs deep algorithmic instincts and parallel programming expertise (Python, C/C++, Fortran, CUDA ecosystems) with hands-on computer vision and embedded GPU work for inspection drones. He gravitates toward paradigm shifts—exploring machine learning’s contrast with traditional programming and the potential of quantum computing—while staying grounded in provable numerical methods. Comfortable collaborating across cultures and disciplines, he has partnered with research groups worldwide and built tooling that spans from raw-image pipelines to end-to-end DL training benchmarks. An intellectually curious problem-solver, he routinely tackles projects outside his comfort zone, turning theoretical insights into production-grade, high-performance code.
13 years of coding experience
8 years of employment as a software developer
University of São Paulo
Portuguese, Spanish, French, English