Nic Eggert
Staff Software Engineer at Stack AV
Inver Grove Heights, Minnesota, United States
Summary
👤
Senior🎓
Top SchoolNic Eggert is a Staff Software Engineer with 14 years of experience building and operating production-grade ML infrastructure, from training neural networks to designing hybrid cloud/on-prem MLOps platforms. At Target he architected GPU compute and serving platforms that supported hundreds of workloads daily, slashed LLM inference costs 70%, and accelerated model delivery from months to under a week. His background spans research-grade algorithm design (PhD in Physics) to pragmatic engineering—implementing Ray cluster provisioning in Go, Kubeflow/Kubernetes orchestration, and FastAPI-based aggregators. An active open-source contributor, Nic has improved distributed training, testing, and core layer functionality in prominent projects like PyTorch Lightning and Keras. Colleagues rely on him to translate complex ML research into reliable, scalable systems used across dozens of teams.
14 years of coding experience
12 years of employment as a software developer
BS Physics (Summa Cum Laude), BS Physics (Summa Cum Laude) at University of MinnesotaPhD Physics, PhD Physics at Cornell University
Github Skills (13)
neural-network10
pytorch10
machine-learning10
deeplearning-ai10
deep-learning10
tensorflow10
python10
data-science10
testing10
web-framework9
artificial-intelligence8
ai8
jax5
Programming languages (13)
Github contributions (5)
Lightning-AI/pytorch-lightning
Aug 2019 - May 2021
Pretrain, finetune ANY AI model of ANY size on multiple GPUs, TPUs with zero code changes.
Role in this project:
ML Engineer
ML EngineerContributions:2 reviews, 24 commits, 23 PRs in 1 year 9 months
Contributions summary:Nic primarily contributed to the testing and development aspects of the PyTorch Lightning framework, focusing on features related to testing, logging, and distributed training configurations. The commits include refactoring test modules, implementing new testing functionalities for multiple dataloaders, and enhancing the logging system to support various frameworks and functionalities. Furthermore, the user addressed issues related to distributed training, especially the ddp_cpu backend.
pythonheadachespytorch-modelsdata-sciencehandling
keras-team/keras
Nov 2015 - May 2016
Deep Learning for humans
Role in this project:ML Engineer
ML EngineerContributions:9 commits, 8 PRs, 44 comments in 5 months
Contributions summary:Nic primarily contributed to the `keras-team/keras` repository by modifying core layers and containers related to deep learning models. Their work included fixing typos in the Lambda layer and allowing it to accept constructor arguments, making it more flexible. They improved performance by adding caching for layer output sizes, and addressed bugs in how nested Sequential containers handle weights. Furthermore, the user refactored and extended the existing functionality.
pythondata-sciencedeep-learningneural-networksmachine-learning
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.