placeholder
Stuart Gentle Publisher at Onrec

GoodData.AI launches AI observability to track enterprise usage, costs and answers

Conceptual illustration of connected data blocks and a laptop representing AI observability

The new capability connects organisation-wide AI performance measures with the execution path behind individual interactions, helping teams investigate quality and cost.

GoodData.AI has launched AI Observability, an extension of its agentic infrastructure designed to show enterprises how AI is being used, how it performs, what it costs and why individual responses behave as they do.

The company says traditional monitoring can reveal a change in usage, quality or cost without explaining the cause. Its new capability links those signals to the steps behind each interaction.

From usage to individual interactions

Teams can monitor query volumes, active users and workspaces, adoption across agents and skills, quality signals, token consumption and cost.

When an answer needs investigation, they can inspect which skills were considered and activated, what knowledge and memory were retrieved, model calls, failures, timing, iteration counts and token usage.

The platform can also analyse conversations for recurring issues and recommend changes to knowledge, semantic models or configuration.

Enterprise teams need more than a dashboard telling them that AI quality or cost changed.

Rosta Striz, Principal Product Manager at GoodData.AI.

Support for teams operating and governing AI

AI Observability runs as a managed workspace using interaction data already generated by the AI. Customers can begin with prebuilt dashboards and create additional views without setting up a separate observability pipeline, according to GoodData.AI.

Deployment is supported in the company’s cloud or on customer-controlled infrastructure.

Engineering teams can use interaction traces to debug agent behaviour, while data teams can investigate how definitions, knowledge and semantic logic affect quality. Product teams can compare adoption and engagement, and compliance teams can use the traceability and audit history to review AI behaviour.

Roman Stanek, CEO and Founder at GoodData.AI, said: “AI becomes much more valuable when enterprises can see how it works and improve it continuously. Observability turns every interaction into evidence: evidence about adoption, quality, cost and the context behind the answer.”

More information is available from GoodData.AI.

Explore AI recruitment suppliers