I'm Maximilian Golub, a Data & Applied Scientist at Microsoft, working on the Azure Hardware Architecture (AHA) team.
Since April of 2020, I've been working on a data science team in Azure, building neat things for big NLP networks. In my spare time, I love taking photos, skiing, 3D printing and reading. Some of my photos are here.
Before Microsoft, I worked at at Mercedes Benz Research and Development in Seattle, where I built Kubernetes enabled full stack applications and worked on distributed machine learning cloud prototypes.
From 2016 to the end of 2018 I was a Master's student at UBC under the supervision of Mieszko Lis and Guy Lemieux, where I researched ways to prune neural networks while training, creating DropBack, achieving state of the art results in pruning WRN-28-10 on CIFAR-10 during training. DropBack enables accelerators like Procrustes that can save energy and improve performance by training with sparsity.
Social Media: * Github