Cube Dev Inc., the creator of an open-source semantic layer that simplifies access to data from disparate systems, is launching an “agentic analytics” platform that uses artificial intelligence to automate data analytics tasks.

With D3, Cube says, it can scale up the productivity of business workers and enable them to explore data independently, without needing to seek help from data professionals first.

The platform introduces the concept of “AI data co-workers” that can automate and enhance analytics tasks, with support for natural language queries, full explainability for every insight, and comprehensive governance.

Cube, which raised $25 million in funding last June, has built an open-source semantic layer that companies can use to analyze business data more efficiently. It addresses challenges around the fact that companies typically use multiple systems to store information, and these must each be accessed using a different application programming interface. In addition, it also overcomes formatting inconsistencies. For instance, one database might only store the times and days that transactions occur, while another might only keep records of the dates, but not the times a sale was made.

Through its semantic layer, Cube organizes information from numerous databases and other systems into a single, consistent format that can be easily accessed using analytics tools via a single API. It eliminates formatting problems and the need to manage multiple APIs.

With Cube’s platform, developers can perform calculations on many different datasets in real time, without any of those hassles. It also provides an in-memory cache that saves the results of frequent calculations, so users don’t have to perform them constantly, meaning lower computing costs.

Now, Cube is adding AI agents into the mix. Agents are one of the hottest trends in AI right now, building on the capabilities of generative AI chatbots that can understand natural language commands. They go further by taking actions on behalf of their users, and can perform various tasks unsupervised. In the case of Cube, its agents are all about automating data analytics.

At launch, Cube D3 features two different AI agents. The first is an AI Data Analyst, which is able to provide self-serve, natural language-driven analytics for any user. Users ask about their data in plain language, and the agent will generate a semantic Structured Query Language query that digs up the insights they need, presenting them in easily digestible, interactive visualizations. In addition, it can also perform tasks such as refining existing reports.

Cube co-founder and Chief Executive Artyom Keydunov told SiliconANGLE the biggest advantage of building AI agents on top of a semantic layer is they gain more context, allowing them to perform tasks for users more effectively.

“LLMs are smart, but they can’t operate without context, especially in the data domain,” he said. “Every company defines metrics like revenue or active customers in a unique way. The semantic layer is a way to structure that knowledge and provide it to LLMs so they have that crucial context.”

There’s also an AI Data Engineer for more advanced users that’s able to automate the development of semantic AI models that can quickly leverage disparate data sources, enabling higher velocity and flexibility for the semantic data layer. Keydunov said that these agents are designed to automate changes in the semantic layer.

Sometimes, the data required for a new analysis is not available in the semantic layer, Keydunov explained. In order to make a change to the semantic layer, the user must know the data domain and possess the SQL and programming skills needed to perform that task, and also be proficient in version control systems.

“That means analysts or business stakeholders often need to request changes to the underlying semantic model from the data engineering team,” Keydunov said.

The problem is that such requests can take weeks or months to process, and that’s what Cube is trying to change with its AI Data Engineer, which can rapidly make and test changes to the semantic layer under the supervision of human workers.

“AI agents can take on the heavy lift of creating SQL queries, writing code to update the semantic layer, and managing all version control and system integrations,” Keydunov explained. “That can facilitate the development of semantic models in organizations and, in general, increase the velocity of analysis. What was taking weeks or months now can be done in a matter of days with less technical personnel.”

While Cube is kicking things off with the above two agents, there are others in the pipeline that it plans to release later this year, including one that will help human data analysts to build and curate interactive data apps that can be shared across an organization and externally with partners. It’s also planning a data science agent, which will be all about generating insights around anomaly detection, predictions and so on.

Crucially, Cube says, its AI data agents operate autonomously within existing enterprise data frameworks, meaning that they preserve all of the same governance controls, trust and transparency of traditional analytics. All of their outputs are explainable, Keydunov said, ensuring full oversight.

Image: SiliconANGLE/Dreamina