Denodo Digs Into Answering the ‘Why’ with DeepQuery

Denodo today unveiled DeepQuery, a new generative AI capability that is now available within its data platform as a private preview. According to Denodo, its new DeepQuery capability leverages the latest in GenAI reasoning models and can not only provide answers about what has happened, but also why something happened.

Denodo is an established Palo Alto, California developer of a data virtualization platform that provides customers with a single plane to manage their data no matter where it resides, which some liken to a data fabric. A key element of Denodo’s platform is a universal semantic layer that helps to unify and standardize data for downstream operational, analytic, and AI use cases.

In recent months, Denodo has added GenAI capabilities to its platform. It has launched an AI-powered assistant and published the Denodo AI SDK, which includes pre-built APIs to help developers connect to popular large language models (LLMs), accessing JSON-formatted data, and initiating retrieval augmented generation (RAG) workflows.

Now the company is going beyond simple chatbots and RAG AI applications with DeepQuery. The new offering, which has been added to the Denodo AI SDK, is designed to help customers build, experiment with, and ultimately integrate so-called Deep Research capabilities into their own agents, copilots, and domain-specific applications.

Standard LLM and RAG workflows are desigend to work with pre-defined questions and modeled data (sasirin pamai/Shutterstock)

The company says DeepQuery goes beyond what’s possible with standard LLMs and RAG applications by enabling organizations to ask “complex, cross-functional questions” that could take a human analyst days to research and answer, which Denodo calls Deep Research.

For instance, questions like “Why did fund outflows spike last quarter?” and “What’s driving changes in customer retention across regions?” could require a human analyst to hunt for multiple pieces of relevant data, figure out how they’re connected, and then stitch them together into a final report. DeepQuery was designed to mimic this type of workflow.

“Instead of retrieving isolated facts, Deep Research breaks down complex questions into smaller, searchable parts; leverages data from multiple sources (e.g., sales, finance, HR, and call center logs); synthesizes insights across data products and documents; and draws conclusions with citations and transparent reasoning paths,” Kevin Bohan, Denodo’s director of product marketing, writes in a blog post today.

“In short, it doesn’t just surface what’s happened–it explains why it happened, what’s driving the trend, and what other factors might be involved,” Bohan wrote. This is the kind of analysis that would normally take a skilled analyst days to complete. Deep Research can do it in minutes.”

Another such question is whether there’s a correlation between the salary of an NBA player and performance metrics. As Denodo explains in a video demo of DeepQuery’s capabilities, “there’s no single table with that answer.” Instead, this type of analysis typically requires an expert to dig into the data, Denodo says.

Denodo’s new DeepQuery capability initiates multiple reasoning agents to answer complex questions (Image courtesy Denodo)

“This is where Denodo’s new DeepQuery functionality comes in,” the company says in the video demo. “When you ask a complex question like this, DeepQuery initiates a multi-step analysis process. It explores multiple tables and runs several queries all automatically to compile a comprehensive report.”

Denodo compares the typical LLM-powered chatbot or RAG workflow to a simple Web search that’s able to provide quick answers and summaries to a large number of relatively simple questions. Bohan writes that the reasoning models available within the Deep Research category, on the other hand, are more geared toward investigations into root causes and patterns; conducting cross-functional analysis that spans data sources; answering open-ended questions that require reasoning, synthesis, or explanation; and answering overall more complex questions.

“AI’s true potential in the enterprise lies not just in generating responses, but in understanding the full context behind them,” stated Angel Viña, the CEO and founder of Denodo. “With DeepQuery, we’re unlocking that potential by combining generative AI with real-time, governed access to the entire corporate data ecosystem, no matter where that data resides. Unlike siloed solutions tied to a single store, DeepQuery leverages enriched, unified semantics across distributed sources, allowing AI to reason, explain, and act on data with unprecedented depth and accuracy.”

DeepQuery is currently available in a private preview mode within Denodo Platform 9.2. Denodo is inviting a handful of organizations to get access to DeepQuery and to collaborate with the Denodo product team by joining its AI Accelerator Program. You can find more information about it here.

Denodo Brings Data Management to Academia, the Cloud

Data Fabric Firm Denodo Raises $336 Million

The post Denodo Digs Into Answering the ‘Why’ with DeepQuery appeared first on BigDATAwire.