Tensorflow and Friends

No, Docker won't save you from driver hell.

For most machine learning workloads, Docker is helpful but not a guarantee for production/staging/development environment parity. Hell, Docker is barely enough to guarantee the reproduction of numerical outputs. The reason is Docker daemon sits on top of the physical operating system for virtualization. Thus, Docker daemon is bounded by its... [Read More]

Managing Multiple AWS and Git profiles

Setting Proper Config To Avoid Pushing Code to the Wrong Place with the Wrong Profile

Being a machine learning engineer is tough nowadays. In addition to managing machine learning pipeline config, we also have to juggle between multiple git profiles and AWS accounts, hoping not to start an expensive training job for work on the personal account, or committing code to work repo using personal... [Read More]