Putting machine learning models into production and keeping them there necessitates investing in well-managed systems to manage the entire data cleaning, training, deployment, and monitoring lifecycle. To keep it running, a set of repeatable and evolvable processes are required. The term MLOps was coined to encompass all of these principles, and the larger data community is working to establish a set of best practices and useful guidelines for accelerating adoption. Demetrios Brinkmann and David Aponte discuss their perspectives on this rapidly changing space and what they have learned from their work building the MLOps community through blog posts, podcasts, and discussion forums in this episode.
A reimagined metadata management platform is required for the modern data stack. Acryl Data’s vision is to provide clarity to your data via its next-generation multi-cloud metadata management platform. Acryl Data, founded by the founders of projects such as LinkedIn DataHub and Airbnb Dataportal, enables delightful search and discovery, data observability, and federated governance across data ecosystems.