Nearly ten years have passed since Amazon Web Services (AWS), the company’s cloud computing subsidiary, unveiled SageMaker, a platform for developing, honing, and implementing AI models. This year, the objective was to streamline, whereas in prior years, AWS concentrated on significantly increasing SageMaker’s capabilities.
AWS introduced SageMaker Unified Studio, a centralized location for locating and interacting with data from within a business, at its re:Invent 2024 conference. In order to assist clients in finding, preparing, and processing data for model building, SageMaker Unified Studio combines capabilities from many AWS services, including the already-existing SageMaker Studio.
Swami Sivasubramanian, AWS’s VP of data and AI, said in a statement, “We are witnessing a convergence of analytics and AI, with customers using data in increasingly interconnected ways.” “The upcoming version of SageMaker combines features to provide users with all the tools they require for generative AI, machine learning model development and training, and data processing, all within SageMaker.”
Customers can publish and share data, models, apps, and other artifacts with their team or a larger organization using SageMaker Unified Studio. The service offers customizable permissions, data security measures, and connections with AWS’ Bedrock model building platform.
SageMaker Unified Studio, specifically Q Developer, Amazon’s coding chatbot, has AI incorporated in. Q Developer can respond to queries such as “What data should I use to get a better idea of product sales?” in SageMaker Unified Studio. or “Create a SQL query to determine total revenue by product type.”
“Q Developer [can] support development tasks such as data discovery, coding, SQL generation, and data integration” in SageMaker Unified Studio, according to an explanation by AWS in a blog post.
AWS introduced SageMaker Catalog and SageMaker Lakehouse, two minor additions to its SageMaker product family, in addition to SageMaker Unified Studio.
Using a single permission model with granular restrictions, SageMaker Catalog enables administrators to create and execute access policies for AI apps, models, tools, and data in SageMaker. In the meanwhile, SageMaker Lakehouse enables connections between SageMaker and additional tools and data housed in corporate apps, data lakes, and data warehouses on AWS.
According to AWS, SageMaker Lakehouse can be used with any tool that complies with Apache Iceberg requirements, which are open source for large analytic tables. If administrators so want, they can implement access controls for all data in all analytics and AI technologies that SageMaker Lakehouse uses.
In a slightly related development, new integrations should improve SageMaker’s compatibility with software-as-a-service applications. Customers of SageMaker don’t need to extract, convert, and load data before accessing it from apps like Zendesk and SAP.
AWS stated that “customers would benefit from a simple way to unify all of this data, as they may have it spread across multiple data lakes and a data warehouse.” “To support use cases like SQL analytics, ad-hoc querying, data science, machine learning, and generative AI, customers can now use their preferred analytics and machine learning tools on their data, regardless of how and where it is physically stored.”