The new managed ML platform is claimed to help companies to more quickly and easily manage AI models
Google Cloud’s Vertex AI to help in faster deployment and maintenance of AI models. (Credit: Gerd Altmann from Pixabay)
Google Cloud has launched a new managed machine learning (ML) platform called Vertex AI that can enable firms to fast track the deployment and maintenance of artificial intelligence (AI) models.
Compared to competitive platforms, Google Cloud said that its new ML platform needs about 80% lesser lines of code for training a model.
Vertex AI can help data scientists and ML engineers with varying levels of expertise in implementing machine learning operations (MLOps) for efficiently developing and handling ML projects across the complete development lifecycle.
Google Cloud cloud AI and industry solutions vice president and general manager Andrew Moore said: “We had two guiding lights while building Vertex AI: get data scientists and engineers out of the orchestration weeds, and create a industry-wide shift that would make everyone get serious about moving AI out of pilot purgatory and into full-scale production.
“We are very proud of what we came up with in this platform, as it enables serious deployments for a new generation of AI that will empower data scientists and engineers to do fulfilling and creative work.”
According to Google Cloud, data scientists are currently struggling with the challenge of manually putting together ML point solutions, which results in a lag time in the development and experimentation of models. This means fewer models end up into production.
For handling such challenges, Vertex AI is said to bring together the services of Google Cloud for building ML under one unified UI and API, for easing the process of developing, training, and deploying ML models at scale.
Vertex AI enables data science and ML engineering teams to access the internally used AI toolkit that powers Google, and deploy more, useful AI applications at a faster rate with the help of new MLOps features.
The new MLOps features include Vertex Vizier, Vertex Feature Store, and Vertex Experiments.
Besides, the new ML platform can help them in eliminating the complexity of self-service model maintenance and repeatability by using MLOps tools such as Vertex Continuous Monitoring and Vertex Pipelines.