Which tool is used for developing agentic systems on Databricks?
Why D is the answer
MLflow with Agent Framework
Agent Framework is the way to build agents on Databricks, and MLflow logs, traces and evaluates them.
In detail
- MLflow and Agent Framework are used to build agentic systems.
- Trace from the start, because it shortens debugging.
Use case
An illustrative scenario, not a real incident.
A team builds an agent with Agent Framework and uses MLflow to log, trace and evaluate it, all in one workflow on Databricks.
See it in different ways
sequenceDiagram participant C as pyfunc chain participant M as MLflow participant U as Unity Catalog participant S as Model Serving C->>M: Log the model M->>U: Register a version U->>S: Deploy to an endpoint
flowchart LR S0["Build with Agent Framework"] --> S1["Log with MLflow"] --> S2["Evaluate"] --> S3["Deploy"]
Key point: Agent Framework builds agents, and MLflow tracks them.
Why the other options are wrong
- A. A SQL dashboard that charts each agent's output
- A dashboard reports on results, and it does not build agents.
- B. Lakeflow Jobs alone
- A job runs work on a schedule, but it is not an agent framework.
- C. A cluster policy alone
- A cluster policy controls compute settings.