Log entry Nº 473 · the stack watch

Bearing SW 225° · November 18, 2026 · 4 min read

Customer Data & CDPSW 225°Watch log

Starburst previews Context Studio for governed AI agent context

Starburst Data opened private preview of Context Studio at its Launch Point conference, targeting a January public preview and Q1 2027 general availability for governed AI agent context.

By Tom Whitfield4 min read772 words

Starburst intros Context Studio to boost trust in AI responses
Fix Nº 473Starburst intros Context Studio to boost trust in AI responses — kewl / pixabay

Starburst Data opened private preview of Context Studio on Wednesday, unveiling the product at its Launch Point virtual user conference. Public preview begins in January, and the vendor expects general availability in the first quarter of 2027.

Context Studio collects the context an AI agent requires and applies it alongside data when the agent queries. It does so without asking customers to build pipelines that move data and business logic into a central location. The product extends Starburst's federated query architecture, so context stays where customers already create and store it.

What does Context Studio actually do?

Three mechanics shape day-to-day use:

  • Any agent connecting to Starburst through a Model Context Protocol server receives the same definitions.
  • Users approve and trace every definition before an agent uses it.
  • A context loop turns agent feedback into proposed improvements; customers approve or decline each one before the change starts influencing other answers.

"Context Studio is a significant new capability that expands on Starburst's federated data processing and data product capabilities to enable customers to identify, reconcile and certify context from across their data estate and provide knowledge that enables trusted agentic decision-making," Matt Aslett, an analyst at ISG, told TechTarget.

Aslett said the federated basis separates Context Studio from the context layers competing vendors are building, which tend to limit agents to context inside their own platform.

"All providers across the analytics and data stack are refactoring their products to provide context to AI agents, [but] … while other platform-specific context environments may be limited to the information available within that platform, Context Studio builds on Starburst's established position as an enabler of federated data processing across a heterogeneous data landscape," Aslett said.

Why Starburst built it

Connecting agents with relevant context remains one of the main reasons most AI initiatives fail, and vendors have moved to close the gap. AWS, Google Cloud and Microsoft have shipped capabilities that let AI tools pull contextually appropriate information in real time. Databricks and Snowflake have done the same, and Alteryx, Informatica and Tableau have released tools designed to connect agents with context.

Starburst's own customers forced the issue, CEO Justin Borgman said. The vendor watched customers try to build agents, and built Aida, its own AI-powered assistant, in parallel.

"As customers started building on top of [Aida], and as we built more of our own agents internally, we all hit the same wall: context," Borgman said.

"Definitions and business logic lived in different tools and none of it worked together, so agents couldn't reach the level of trust needed to move from pilot to production. That's why we built Context Studio to harvest and serve the context you already have."

Aida, introduced in April and generally available in May, already runs on Context Studio. The assistant draws on governed definitions to answer with context federated across an enterprise's various sources.

What analysts say is missing

Devin Pratt, an analyst at IDC, said the significance of Context Studio is that it adds meaning to the data Starburst users can already query wherever it lives. That meaning lets agents find and operationalize context, which makes responses more accurate and trustworthy.

"Agents answer from definitions, [and] Context Studio helps make sure they're the right ones," Pratt said.

Pratt also said Starburst stands apart by applying definitions where data is created and stored, rather than making customers build costly and complex pipelines into a centralized environment.

"Most approaches ask enterprises to move their definitions in," Pratt said. "Starburst brings context to where the data already lives."

Aslett said Context Studio covers business terms, rules and metrics, plus business domains and query histories. He also said reasoning and planning, along with the outcomes of previous decisions and actions, remains a work in progress across the vendor category and would be a logical addition as development continues. Pratt suggested adding the same versioning capabilities Starburst applies to data products.

"Trust is traceable -- every definition should show its source, its approver and its history," Pratt said.

During the preview, Starburst said it will monitor how well Context Studio surfaces and addresses conflicting definitions, whether it gives different agents consistent answers, and whether it preserves the access controls customers already have in place.

What does this change for data teams?

Practitioners get one governed place to approve, trace and version the definitions agents read, without relocating their data, plus a shipped use case in Aida. The open questions are timing: public preview in January, general availability in the first quarter of 2027, and reasoning and decision-outcome history still on the roadmap.

Plot points

  1. Context Studio entered private preview at Starburst's Launch Point virtual user conference on Wednesday
  2. Public preview begins in January; general availability is expected in the first quarter of 2027
  3. Agents connecting via a Model Context Protocol server receive identical, user-approved definitions
  4. Starburst's Aida assistant launched in April and reached general availability in May
  5. IDC analyst Devin Pratt said: "Most approaches ask enterprises to move their definitions in. Starburst brings context to where the data already lives."

via techtarget.com (Source)

Filed under

  • starburst
  • context-studio
  • ai-agents
  • data-governance
  • federated-data
Share this article:

More from Tom Whitfield

Tom Whitfield

Show full bio

News editor covering money, funding and market shifts around marketing technology at Martech Beacon.

37 articles

Neighbouring fixes

‹ Previous article