Delivery Layer reposted this
With traditional software you can "set it and forget it," but sadly data products rarely just keep working after you build them. You need to set it, regularly retrain it, fix the data pipelines, create new features, adjust the model when you add a new option to the onboarding survey, etc... Maintaining data products is as challenging as building them. One of the most important parts of product management for data products is preparing your company for when this happens.
Andrew C. Madson
2mo
100% - the difference between a data asset and a data product is the continual process that builds, monitors, and improves them.
Forget about historical KAHN-vention. You need to KAHN-tiniest to maintain your data and AI products more than you do traditional products. See Solomon Kahn… a Solomaintenance-Man as well as a Solominnovation-Man of AI… and among Top Data People. You can be a top data person too.