Stardog, a Washington, DC area startup known for assisting government agencies in managing their data, recently unveiled Voicebox, a conversational layer powered by LLM. This innovative solution, integrated into Stardog’s flagship platform, aims to facilitate access to business insights by allowing users to ask questions in everyday language and receive answers based on enterprise data, without requiring any technical expertise.
The integration of large language models into data management processes is a growing trend in the industry. Stardog joins other players in the data ecosystem, such as Kinetica, Databricks, and Dremio, in leveraging these models to streamline data workflows. According to Kendall Clark, co-founder and CEO of Stardog, this development has a significant impact on competitiveness and profitability, as it eliminates the longstanding challenge of universal access to relevant data.
“Self-serve analytics is no longer the exclusive preserve of technical folks who’re able to program.” – Kendall Clark
A Breakthrough in Data Querying
Voicebox operates similarly to a natural language processing (NLP) bot, where users can simply type their queries to obtain answers. However, this AI-powered layer goes beyond retrieving raw data by directly interacting with Stardog’s Enterprise Knowledge Graph.
The Enterprise Knowledge Graph is a comprehensive tool that connects to all data sources within an organization, enriches the information with contextual relevance, and transforms it into human and machine-understandable knowledge. Voicebox leverages this graph to provide accurate and timely answers based on trusted enterprise data.
“It all works by taking in natural language and using our models to turn human intent into structured graph queries that Stardog executes.” – Kendall Clark
With Voicebox, users no longer need to write complex queries for the knowledge graph platform. The introduction of “data democratization” ensures that all knowledge workers can access and utilize the platform, while maintaining data integrity and accuracy. Stardog’s prompt engineering and agent techniques play a crucial role in delivering safe, trusted, and contextually relevant querying experiences.
Expanding Accessibility and New Offerings
Stardog has made significant strides in expanding the accessibility of its Voicebox solution. Early access has been granted to numerous existing customers and new prospects, including businesses in manufacturing and pharmaceuticals. Their feedback will help shape the implementation of LLM (Large Language Models) into data projects and expedite the introduction of new features.
Stardog is actively working on integrating SMS and WhatsApp support for Voicebox, ensuring seamless integration into users’ digital workflows. Additionally, there are plans to explore voice prompts in the future, although no official timeline has been announced.
While the company remains tight-lipped about specific enterprise clients using Voicebox, Stardog’s Enterprise Knowledge Graph has attracted prominent organizations like Boehringer Ingelheim, Schneider Electric, NASA, and the Department of Defense. According to Forrester, the platform offers a significant ROI of 320% and total benefits exceeding $9.86 million over three years.
Although Stardog has an edge with its knowledge graph technology, other companies are also leveraging large language models to simplify data access. Kinetica, Snowflake, Databricks, and Informatica have all introduced their own AI-powered solutions aimed at enhancing data handling and querying processes.
