Request a demo

Transformations

Python and SQL you can read, on the node that runs it.

Describe the step in plain language. The agent writes the script from the columns already on the graph, attaches it to the node, and holds it for review before it runs.

Request a demoHow it works

The script names fields that exist

Written from real columns

The script names fields that exist

The churn score was written against order_date, attempted_at, and customer_id because the upstream nodes emit them. Each run records the rows and columns it produced, step by step.

  • Python, with SQL where SQL is clearer
  • The instruction that produced it, kept on the node
  • Rows in and rows out for every step of the script

Transformations in detail

  • 01

    Readable by default

    Every transformation is a script you can open, copy, expand, and edit.

  • 02

    Intent stays attached

    The plain-language instruction lives next to the code it produced.

  • 03

    Review before run

    A new script waits for a person before the run is allowed to execute it.

  • 04

    Edit when you disagree

    Change the script yourself. The node runs the version you approved.

  • 05

    Test on samples

    Test mode dry-runs nodes against sample data without touching downstream artifacts.

  • 06

    Summaries, not guesses

    Each run records the output row count and column list for the next node to build on.

Related capabilities

All capabilities

Ready to build a pipeline you can inspect?

Describe the outcome, inspect what the agent proposes, and run it when you trust it.

Request a demoLog in
How it worksConnect, shape, run, and ask, step by step.API referenceCreate an API key and call Ordiva from your own code.