Cloth information brokers work with current OneLake implementations, giving them a base set of knowledge to make use of as context in your queries. Alongside along with your information, they are often fine-tuned utilizing examples or be given particular directions to assist construct queries.
There are some conditions earlier than you may construct an information agent. The important thing requirement is an F64 or greater consumer, together with an acceptable information supply. This is usually a lake home, an information warehouse, a set of Energy BI semantic fashions, or a KQL database. Limiting the sources is sensible, because it reduces the chance of dropping the context related to a question and retains the AI grounded. This helps make sure the agent makes use of a restricted set of recognized question sorts, permitting it to show your questions into the suitable question.
Constructing AI-powered queries
The agent makes use of consumer credentials when making queries, so it solely works with information the consumer can view. Function-based entry controls are the default, holding your information as safe as doable. Brokers’ operations must keep away from leaking confidential info, particularly in the event that they’re to be embedded inside extra advanced Azure AI Foundry agentic workflows and processes.
