Continuously Improving Recommender Systems for Competitive Advantage Using NVIDIA Merlin and MLOps

Originally published at: Continuously Improving Recommender Systems for Competitive Advantage Using NVIDIA Merlin and MLOps | NVIDIA Developer Blog

Recommendation systems must constantly evolve through the digestion of new data or algorithmic improvements of the model for its recommendations to stay effective and relevant. In this post, we focus on how NVIDIA Merlin components fit into a complete MLOps pipeline to operationalize a recommendation system, and continuously deliver improvements in production

With this post, we aim to address how developers can build a performant recommendation system with NVIDIA Merlin employing MLOps tools and practices. This solution and blog post is aligned with the GTC Spring '21 talk “Gain Competitive Advantage using ML Ops: Kubeflow and NVIDIA Merlin and Google Cloud”, please check that out for more context, as well as for the demo video!

Hello, I’d like to check if there’s some published work on implementing the complete Merlin pipeline for a recommender system in production using AWS. I have found a blog that details that implementation on Google Cloud and I wonder if there’s a published implementation on AWS.

Thanks