Tiago Antunes
Network Engineer at Unidade de Computação Científica Nacional - FCT
Lisbon, Lisbon, Portugal
Summary
🤩
Rockstar🎓
Top SchoolTiago Antunes is a Network Engineer with nine years of hands-on experience designing, commissioning and supporting optical transmission and IP networks, currently based in Lisbon. He has deep DWDM/OTN expertise from roles at Coriant, Nokia and ZTE/Altran and now applies that knowledge to national research computing networks at Unidade de Computação Científica Nacional. Beyond field and operational work, he has taught computer networks at university level and led integration, testing and 100Gb/s optical transmission projects, blending practical deployment skills with rigorous test discipline. Unusually for a telecoms engineer, he also contributes to high-performance ML tooling—optimizing CUDA kernels for a PyTorch Mixture-of-Experts library—demonstrating strong low-level performance coding alongside network systems expertise.
9 years of coding experience
11 years of employment as a software developer
Instituto Superior Técnico
Portuguese, English
Github Skills (8)
cuda10
pytorch10
machine-learning10
parallel-computing9
cprogramming-language9
performance-optimization9
c-language9
deep-learning8
Programming languages (8)
Github contributions (5)
laekov/fastmoe
Mar 2021 - Jun 2021
A fast MoE impl for PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:10 reviews, 15 commits, 7 PRs in 2 months
Contributions summary:Tiago primarily focused on optimizing the `fastmoe` library for PyTorch, a library for implementing Mixture-of-Experts (MoE) models. They made significant contributions by modifying and improving the CUDA kernels, specifically for the MOELinear layer, to incorporate bias calculations directly. The user also introduced new CUDA kernels for column reduction, enhancing performance. Furthermore, they added a test suite for different data types, ensuring cross-compatibility within the kernels.
pytorchmoemixture-of-expertsimplpytorch-lightning
TiagoMAntunes/TiagoMAntunes
Jul 2020 - Nov 2022
Contributions:8 commits, 6 pushes, 1 branch in 2 years 4 months
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.