Google Makes Vertex AI Generally Available to Streamline Machine Learning Workflows
Google launches Vertex AI as a fully managed platform that unifies separate machine learning tools into a single interface to speed up model development and deployment.
Google announces the general availability of Vertex AI, a managed machine learning platform designed to speed up the development and maintenance of artificial intelligence models. By combining previously separate Google Cloud services into a single user interface and API, the platform eliminates the need for data scientists to manually stitch together disjointed machine learning point solutions.
Developers use Vertex AI to create machine learning pipelines that train and evaluate models using Google Cloud algorithms or custom training code across image, video, text, and tabular data. They then deploy these models using scalable managed infrastructure for both online and batch use cases, which allows them to move models from experimentation to production much faster.
The unified platform provides access to the same AI tools Google uses internally, including prebuilt containers for TensorFlow, XGBoost, and Scikit-learn. Additional features like Vertex Feature Store, Vertex Experiments, and the upcoming Vertex ML Edge Manager help practitioners share machine learning features, increase experimentation rates, and build models directly on edge devices.