Product
A job is a thing you can point at.
Attach a notebook or a SQL script, choose a trigger, and it runs. No separate orchestration repo, no DAG written twice.
Triggers
Run it manually, on a CRON schedule, or as a dependency of another job finishing.
Runs you can read
Every run records its inputs, outputs and the compute it used, which is the same record the catalog and the invoice read from.
Promotion between domains
A job written against finance_dev moves to finance_prod without rewriting paths, because the domain and environment are part of the catalog name rather than hardcoded in the code.
See it against your own data
We will walk through your workloads and show what runs where.