Starburst 增加上下文支持,赋予 AI 智能体对业务数据定义的持续访问能力

媒体报道 SiliconANGLE AI 原文语言:英文

Starburst Data Inc. 推出业务上下文管理功能,正在将其数据联合平台进行扩展。

全文

UPDATED 08:00 EST / NOVEMBER 18 2026

AI

Starburst adds context support to give AI agents consistent access to business data definitions

by Paul Gillin

Starburst Data Inc. is expanding its data federation platform into business context management with the introduction of Context Studio, a new capability designed to help artificial intelligence agents produce more accurate and trustworthy answers from information distributed across enterprise systems.

Context Studio collects business definitions from existing data catalogs, business intelligence tools, metric stores and documents, making them available alongside data when AI agents respond to prompts. The system is intended to eliminate conflicting definitions that can cause AI agents to produce incorrect results.

The new capability addresses a problem the company encountered as customers began deploying its Aida natural-language data assistant in production environments, said Starburst co-founder and Chief Executive Justin Borgman.

“We’ve found that as you get into production, the biggest issue is context because you need to trust the answers the agent gives back,” he said.

The challenge is particularly acute in large organizations where different departments may use different definitions of common business metrics, such as revenue or active customers. Without a consistent understanding of those terms, AI models can produce plausible but inaccurate answers.

“When those definitions conflict, the [large language model] doesn’t know any better; it’s just going to guess,” Borgman said. “That’s what leads to bad answers and an erosion of trust.”

Data products

Context Studio builds on Starburst’s existing data products architecture, which allows organizations to create curated views spanning multiple data sources. Those products incorporate business definitions, metadata and access controls that can be applied consistently when agents retrieve information.

Unlike conventional semantic layers that require organizations to consolidate definitions within a single platform, Context Studio federates existing definitions without requiring customers to move or rebuild their underlying data.

Built-in approval workflows resolve conflicting definitions, and a feedback mechanism is provided through which agents can propose improvements. Human reviewers approve changes before regression testing checks their impact on other queries.

Borgman characterized the expansion as an evolution of Starburst’s longstanding strategy of querying distributed data without centralizing it.

“We think context is going to live everywhere, just like we think data will live in multiple places,” he said. “Our approach treats [context] as a first-class principle in our design.”

Starburst hasn’t yet disclosed quantitative evidence of Context Studio’s impact on AI accuracy or productivity. The company said it is collecting benchmark data from private preview customers.

The company is also emphasizing traceability to increase confidence in AI-generated answers. Context Studio records which data sources agents access, the SQL queries they generate and the information used to produce responses. Access controls remain attached to underlying data products.

“Context is going to increase the probability of a good answer, but there’s also a need to trace how it got to that answer,” Borgman said.

An agent can interview domain experts to capture business knowledge and convert it into draft data products. Starburst is also extending its Aida data assistant to answer questions across multiple data products and introducing an agent that automatically generates dashboards.

Other platform enhancements include Mission Control for monitoring usage and costs, automated Apache Iceberg maintenance, a bring-your-own-cloud deployment option and graphics processing unit-accelerated querying. Starburst claims its recently introduced Torin query engine delivers up to sevenfold performance improvements in company benchmarks.

Context Studio is available in private preview for Starburst Galaxy and Starburst Enterprise. A public preview is planned for January, with general availability expected in the first quarter of 2027. Pricing hasn’t been finalized.

Image: Starburst Data

A message from John Furrier, co-founder of SiliconANGLE:

支持我们的使命,通过参与theCUBE社区保持内容的开放和免费。加入theCUBE校友信托网络,在这里技术领袖们相互联系、共享情报并创造机会。

超过1500万观众观看theCUBE视频,推动AI、云计算、网络安全等领域的对话。

超过11.4千名theCUBE校友——与超过11,400位塑造未来的技术和商业领袖通过独特的信任网络联系。

您是AWS客户吗?通过在我们的Marketplace门户页面购买AWS服务来为SiliconANGLE提供财务支持,链接:https://siliconangle.com/aws-marketplace/

关于SiliconANGLE Media

SiliconANGLE Media是数字媒体创新的公认领袖,融合突破性技术、战略洞察和实时受众互动。作为SiliconANGLE、theCUBE Network、theCUBE Research、CUBE365、theCUBE AI和theCUBE SuperStudios的母公司——在硅谷和纽约证券交易所设有旗舰地点——SiliconANGLE Media在媒体、技术和AI的交叉点运营。

由技术愿景家John Furrier和Dave Vellante创立,SiliconANGLE Media构建了一个动态的行业领先数字媒体品牌生态系统,覆盖1500多万精英技术专业人士。我们全新的专有theCUBE AI Video Cloud在观众互动方面开创先河,利用theCUBEai.com神经网络帮助技术公司做出数据驱动的决策,并保持行业对话的前沿。

正文为 AI 译文。

参考来源

← 返回每日AI报告