Summary
Filippo Arcadu is a Computer Vision Engineer and tech lead based in Zurich with 11 years of experience applying advanced ML to medical imaging and scientific problems. He holds a PhD from ETH Zürich and has driven production-grade deep learning work at Roche—spanning 3D U-Net segmentation, Mask R-CNN, multiple-instance learning and video-level disease classification—and now applies that expertise at Meta. His background in numerical optimization and inverse problems informs a principled approach to imaging pipelines, from ultra-fast reconstruction algorithms for X-ray tomography to CUDA-accelerated model training. Comfortable across Python/C++ stacks and frameworks (PyTorch, TensorFlow, OpenCV) and HPC environments (Slurm, CUDA), he bridges research rigor with deployment-ready systems. Notably, his trajectory from physics and radiotherapy dose optimization to large-scale computer vision gives him an uncommon blend of theoretical depth and practical impact in medical and industrial imaging.
11 years of coding experience
1 year of employment as a software developer
Bachelor’s Degree, Physics, 110 magna cum laude, Bachelor’s Degree, Physics, 110 magna cum laude at Università degli Studi di Firenze
Doctor of Philosophy (Ph.D.), Department of Information Technology and Electrical Engineering, Doctor of Philosophy (Ph.D.), Department of Information Technology and Electrical Engineering at ETH Zürich
English, German, French, Italian