Giuseppe Franco
Member Of Technical Staff at AMD
Cologne, North Rhine-Westphalia, Germany
Summary
🤩
Rockstar🎓
Top SchoolGiuseppe Franco is a machine learning-focused software engineer and biomedical researcher with 10 years of experience blending ML techniques with practical biomedical applications. After top-ranked degrees in Bionics Engineering (110/110 cum laude) and Biomedical Engineering, he contributed to Xilinx’s Brevitas library—extending quantized neural network support and PyTorch JIT integration—before moving into biomedical research at the University of Warwick. He then transitioned to industry roles at AMD, progressing from Senior SDE to Member of Technical Staff, where he applies model optimization and systems-level thinking to production environments. Comfortable at the intersection of research and engineering, he has a track record of shipping robust quantization features and integrating them into real-world workflows. Based in Cologne, he brings a rare combination of academic rigor and hands-on open-source impact that accelerates ML deployment in constrained, biomedical-grade systems.
10 years of coding experience
9 years of employment as a software developer
Bachelor of Engineering - BE, Bioengineering and Biomedical Engineering, 110, Bachelor of Engineering - BE, Bioengineering and Biomedical Engineering, 110 at Politecnico di MilanoMaster of Science - MS, Bionics Engineering, 110/110 cum laude, Master of Science - MS, Bionics Engineering, 110/110 cum laude at Università di Pisa
English, Italian, French
Github Skills (9)
pytorch10
machine-learning10
deep-learning10
onnx9
faster-rcnn8
compiler8
mask-rcnn8
compile8
jit8
Programming languages (4)
Github contributions (5)
Xilinx/brevitas
Jan 2020 - Jan 2023
Brevitas: neural network quantization in PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:2 releases, 604 reviews, 117 commits in 3 years
Contributions summary:Giuseppe primarily contributed to the core Brevitas library, focusing on neural network quantization in PyTorch. Their work involved the merging and modification of quantization-related functions, specifically for PrescaledIntQuant and related classes. The user also implemented and refined various aspects of the quantization process, including integration with Pytorch's JIT compiler and improved testing procedures. Furthermore, the user added support for new layers like QuantConv1d and QuantConvTranspose1d, demonstrating significant contributions to extending the Brevitas library.
pytorchxilinxdeep-learningquantization-aware-trainingneural-networks
Giuseppe5/brevitas
Nov 2019 - Apr 2025
Training-aware quantization in Pytorch
Contributions:14 PRs, 1949 pushes, 469 branches in 5 years 5 months
pytorchquantization-aware-trainingtrainingtorchvisionwandb
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.