Zachary Mueller
Head Of Developer Relations
Columbia, Maryland, United States
Summary
🤩
Rockstar🎓
Top SchoolZachary Mueller is a developer relations leader and ML engineer with eight years of experience building and scaling developer-focused AI tools and communities. He rose from practical machine learning and computer vision research at the University of West Florida to technical leadership roles at Hugging Face and Lambda, contributing core features to widely used projects like Accelerate and Transformers. As a recognized fastai "World Expert" and active fastai forums admin, he blends deep open-source engineering—checkpointing, seed management, and no-trainer examples—with clear developer-facing content and community growth. He also founded Mueller Technologies to teach practitioners how AI systems work and to apply ML toward environmental problems, reflecting his dual interests in education and ecology. Notably, his open-source work spans low-level library hygiene and impactful features (e.g., Accelerate checkpointing and Transformers example consistency), demonstrating both attention to detail and production-grade distributed training expertise. Based in Columbia, Maryland, he thrives at the intersection of developer tooling, model engineering, and community-driven learning.
8 years of coding experience
7 years of employment as a software developer
Minor Computer Science, Minor Computer Science at University of West Florida
Github Skills (28)
pytorch10
hub10
acc10
python10
jupyter10
image-classification10
huggingface-hub10
acceleration10
machine-learning10
accelerator10
fastai10
huggingface10
deep-learning10
documentation-generator9
data-augmentation9
Programming languages (11)
Github contributions (5)
muellerzr/Practical-Deep-Learning-for-Coders-2.0
Oct 2019 - Nov 2021
Notebooks for the "A walk with fastai2" Study Group and Lecture Series
Role in this project:
Data Scientist
Data ScientistContributions:250 commits, 18 PRs, 245 pushes in 2 years 1 month
Contributions summary:Zachary appears to be contributing to a practical deep learning project focused on image classification. The commits involve uploading and adding notebooks that demonstrate the use of custom datasets and the DataBlock API to classify images. They have demonstrated an understanding of dataset creation, labeling, and data augmentation techniques within the fastai2 framework.
pytorchwalkmachine-learningnotebooksstudy-group
huggingface/accelerate
Mar 2022 - Jan 2023
🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support
Role in this project:Back-end Developer
Back-end DeveloperContributions:27 releases, 1370 reviews, 309 commits in 10 months
Contributions summary:Zachary's commits focus on adding checkpointing capabilities and random seed functionality to the Hugging Face Accelerate library, a tool for distributed training. Their contributions include the implementation of saving and loading model states, optimizers, and random number generator states, as well as the ability to manage and manipulate data samples for optimal training and the ability to set the seed with randomness from inside Accelerate. These additions support the library's goal of simplifying and accelerating PyTorch model training.
pytorchpytorch-modelsmixed-precisiondeep-learninggpu
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.