Joseph Mayer
Hardware Engineer at Microsoft
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolJoseph Mayer is a hardware engineer based in Seattle with 3+ years of professional experience building high-performance FPGA and accelerator firmware, currently contributing to Microsoft’s BrainWave and DeepSpeed efforts. He combines academic rigor (MSc in Electrical Engineering, 3.85 GPA) and hands-on firmware development at institutions like CERN and UW, where he implemented FPGA designs for the ATLAS Pixel Detector and ITk upgrades. At Microsoft he bridges hardware and ML software, contributing to the DeepSpeed open-source project by improving distributed training optimizations, debugging engines, and hardening tests—demonstrating a rare cross-domain fluency in both FPGA hardware and ML training infrastructure. Earlier roles in finance and academia sharpened his production-focused design skills and mentoring ability. Colleagues can expect a pragmatic engineer comfortable moving between low-level HDL, system integration, and ML-facing optimizations.
3 years of coding experience
5 years of employment as a software developer
Master of Science (M.Sc.), Electrical Engineering, 3.85, Master of Science (M.Sc.), Electrical Engineering, 3.85 at University of WashingtonBachelor of Science (BS), Computer Engineering, 3.85, Bachelor of Science (BS), Computer Engineering, 3.85 at Saint Louis University
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (17)
algorithm10
pytorch10
python10
optimizers10
machine-learning10
deep-learning10
optimisation10
optimization10
parallelization9
data-parallel9
gpu9
data-parallelism9
zeroth8
inference8
zero-code8
Programming languages (1)
Github contributions (5)
deepspeedai/DeepSpeed
Oct 2022 - Jan 2023
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Role in this project:
ML Engineer
ML EngineerContributions:2 releases, 50 reviews, 48 commits in 3 months
Contributions summary:Joseph primarily contributed to the DeepSpeed library, focusing on enhancements related to optimization and model training. Their work involved fixing bugs, refactoring and extending the DeepSpeed engine, and adding configuration options for gradient accumulation data types. They also updated documentation and addressed issues in the testing framework to ensure the proper functioning of the library, demonstrating a focus on improving the usability and correctness of the software for machine learning workloads.
billion-parametersfinetuningtrainingmixture-of-expertszero
jglaser/DeepSpeed
Aug 2023 - Oct 2023
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
Contributions:2 pushes in 1 month
pytorchdeepspeeddeep-learningeffectiveoptimization
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.