Google today announced a new cloud service that’s designed to make it easier for companies to create custom machine learning algorithms for processing images. Called Cloud AutoML Vision, the system allows developers to upload a bunch of images to Google’s cloud and receive a custom model in return.

It’s based on Google’s research into training machine learning models to construct models that perform particular tasks well. In theory, companies should be able to feed the system a set of sample images and, within a day, get back an automatically trained model that’s optimized for their specific data.

Cloud AutoML, which will eventually expand beyond images, is supposed to help bridge the gap between the companies that need custom machine learning tools and the handful that are able to pay top dollar for the technical talent needed to implement those tools. By creating a system that can build its own models, Google is hoping to make advanced machine learning systems more accessible to everyday developers who don’t necessarily have all the skills in-house.

Machine learning systems that create custom models, especially in the vision space, are nothing new. What sets Google’s apart is its ability to create a bespoke model without human intervention, something that the company helped pioneer. Salesforce is working in a similar space with its own AI products, but it’s difficult to determine how similar the two are at this juncture.

Google’s system is based on supervised learning, which means companies need to provide a bunch of data with labels to tell Cloud AutoML what it’s looking at and help it create a model. Google has a crowdsourced system available that will use humans to generate labels for unlabeled data, but companies can also provide their own labeled data when they first set the system up.

Right now, Cloud AutoML Vision is available in closed alpha testing, and Google will be working with each company that signs up to determine when the service would be right for them. The tech titan has already tested the service with clients like Disney and Urban Outfitters, with positive results.

This new service goes beyond Google’s existing portfolio of cloud machine learning services, which are aimed at either assisting with the broadest number of applications or helping existing machine learning practitioners streamline the process of plying their craft. If the company is successful, it could gain a major advantage in the cloud wars, since potential customers are looking for ways to get machine learning initiatives off the ground without Google-level technical talent.