Datadog Inc. senior FinOps analyst Deeja Cruz emphasized at FinOps X 2026 that effective AI cost management relies on foundational cloud practices: strict data tagging and clear model governance. Without high-quality attribution tags, enterprises cannot allocate AI spend or identify optimization opportunities, risking unchecked costs as workloads scale.
The necessity of data tagging
Organizations moving from traditional cloud infrastructure to AI workloads must maintain rigorous metadata standards. Cruz told theCUBE that poor tagging immediately breaks the ability to answer executive questions about where money is going.
"The biggest takeaway that I can give is don't neglect your tags," Cruz said. "Having good tagging on your data will unlock your ability to allocate it and be able to answer questions that executives are asking."
This discipline allows finance and engineering teams to track exactly which models or features drive specific costs.
Practitioner-led AI and model governance
Cruz highlighted how non-technical staff can directly drive cost savings using generative AI tools. In one example, a colleague without a development background identified a storage bucket configuration issue, used a large language model to generate the necessary code changes, and submitted a pull request that resulted in material savings days later.
"I would encourage all FinOps practitioners to get really comfortable with these tools," Cruz said. "Take your domain expertise and use these tools to deliver value for the organization faster."
On the governance side, Datadog avoids defaulting to the most expensive AI models. Instead, the company evaluates which specific model fits a given workload. This requires a clear division of labor, with the FinOps team handling forecasting and attribution, while an internal developer experience team manages governance tooling and feedback.
Why this matters for managers
Leaders must establish clear ownership frameworks before AI spending scales out of control. Cruz noted that this accountability structure mirrors how cloud ownership evolved, requiring defined leads and support roles across finance, engineering, and security. Managers who formalize these cross-functional responsibilities early can prevent cost overruns. This clarity also serves as a practical foundation for any AI Learning Path for Finance Managers aimed at enforcing sustainable budget control.
Your membership also unlocks:
