Citizen developers need a low-code transformation. Engineers need complex PySpark logic. Which pairing is right?
Why D is the answer
Dataflow Gen2 for the first, notebooks for the second
Dataflow Gen2 is low-code, and notebooks give engineers full control over PySpark.
In detail
- Dataflow Gen2 is low code.
- Notebooks suit complex PySpark.
- Both tools can write to the same store.
Use case
An illustrative scenario, not a real incident.
An analyst cleans a supplier file in Dataflow Gen2, while engineers write a fraud feature in a PySpark notebook. Both feed one lakehouse.
See it in different ways
flowchart TD
A{"Who builds it?"} -->|"Low-code users"| B["Dataflow Gen2"]
A -->|"Engineers, complex logic"| C["Notebook with PySpark"]
A -->|"SQL developers"| D["T-SQL in a warehouse"]mindmap
root((Fabric tools))
Pipeline
Copy
Schedule
ForEach
Notebook
Python
Complex logic
Dataflow Gen2
Low code
Triggers
Schedule
Eventflowchart TD
Q{"What is the work?"}
Q -->|"Low-code cleaning"| A0["Dataflow Gen2"]
Q -->|"Complex code"| A1["Notebook"]Key point: Match the tool to the person and the logic.
Why the other options are wrong
- A. T-SQL only for both groups, using stored procedures
- T-SQL alone does not cover PySpark logic.
- B. Notebooks for both groups, using shared templates
- Notebooks need code that citizen developers may not write.
- C. KQL for both groups, using saved queries and functions
- KQL is not a general transformation tool for both.