Too big to deploy: How GPT-2 is breaking production serverstowardsdatascience.com·2 pts·calebkaiser·0
We built our ML platform on AWS–and why that may have been a mistaketowardsdatascience.com·4 pts·calebkaiser·1
Why we’re writing machine learning infrastructure in Go, not Pythontowardsdatascience.com·130 pts·calebkaiser·74
Why we’re writing machine learning infrastructure in Go, not Pythontowardsdatascience.com·5 pts·calebkaiser·0
A list of beginner-friendly NLP projects–using pre-trained modelstowardsdatascience.com·50 pts·calebkaiser·1