Google has announced the launch of its new AI Hub, enabling firms to leverage plug-and-play machine learning (ML) content and circumvent a deficit in expertise in the workforce
In a blog post, Google’s Hussein Mehanna said that among the 20mn software developers around the world there are only 2mn data scientists.
With the AI Hub, firms can now mitigate the difficulty caused by this gulf in ML expertise by having access to a range of strong ML resources developed by Google Cloud AI, Google Research, and other dedicated teams.
The Hub also enables enterprises to upload and share their own ML solutions to a private, secure platform.
This collaborative scope means that firms can use the AI Hub as a platform for the development of ML solutions from the prototype phase through to production.
Kubeflow Pipelines, released alongside the AI Hub to facilitate this development, serves as a means for enterprises to construct and package ML resources to maximise their benefit to as many internal users as possible.
Pipelines is a component of the popular opensource project Kubeflow which compiles ML code in a similar manner to the building of an app, boosting reusability.
“Kubeflow Pipelines provides a workbench to compose, deploy and manage reusable end-to-end machine learning workflows, making it a no lock-in hybrid solution from prototyping to production,” Mehanna said in the post.