Job

A Job in Polysync is a template for a unit of work that can be executed on a connected Platform — for example a single Azure Data Factory pipeline, a Databricks notebook, an Azure Function, or a Google Cloud Function. A Job captures:

The Platform it runs on.
The Job type (e.g., ADF Pipeline, Databricks Notebook) — see Supported job types below.
The external identifier of the artefact (pipeline name, notebook path, function name, …).
The parameter schema — the names, data types, and default values of the parameters the artefact accepts.
Job-level metadata — name, description, tags.

Jobs do not run on their own. To execute a job you create one or more Tasks from it. Each Task pins concrete parameter values, schedules, dependencies, retry behaviour, and contention profiles.

Editor tabs

Tab	What you configure
General	Job name, Platform (locked after save), Job Type, Platform Job picker (pre-fills settings from the platform artefact), External Id.
Parameters	The parameter schema — names, data types, and default values that Tasks will inherit.
Settings	Job-level attribute values (e.g., connection or execution settings specific to the job type).

How Jobs are created

Import from a Platform — recommended. Open the Platform editor, click Get Pipelines, pick the pipelines you want, and Polysync will create or update the Jobs (and their parameter schemas) for you. The AI Copilot can do this end-to-end via list_platform_pipelines → import_platform_jobs.
Via the Job editor — create a new Job, select the Platform and Job Type, then pick a Platform Job from the dropdown. Polysync fetches the artefact's parameter list and pre-fills the Parameters and Settings tabs automatically. Adjust the External Id and parameter defaults as needed and save.
Manual — leave the Platform Job picker empty and fill in the external identifier and parameter schema by hand.

Configuring a Job

Open a Job in the editor (or ask the AI Copilot via configure_job) to update its display name, description, parameter defaults, or any other attribute. Use the Sync Parameters button to re-fetch the parameter list and settings from the platform after the artefact has changed — this keeps the schema in sync without recreating the Job.

Creating Tasks from a Job

Use the Create Task button in the Job editor (or the AI Copilot's create_task_from_job tool) to spin up a Task. You provide a Task name and the Task starts as a fresh instance of the Job's parameters; you then override values, add a schedule, and link contention profiles.

Execute Now (ad-hoc execution)

The Execute Now button runs the job directly on the Platform without creating a persistent Task. A parameter-override dialog lets you change any value for that single execution. Use this to:

Validate a newly configured Job before creating Tasks.
Trigger a one-off run without disturbing the Task schedule.
Test parameter combinations interactively.

Execution is monitored live in the editor — the button changes to a spinning Running indicator while the job is active.

Supported job types per platform

Each platform exposes one or more job types. The job type determines how Polysync discovers the artefact, how it binds parameters at run time, how it tracks the run, and whether output parameters and cancellation are supported. Each job type has its own page documenting parameter shape, output handling, monitor URL, and troubleshooting tips.

Best practices

Name Jobs after the artefact they wrap (e.g., adf-sales-etl) so they're easy to find.
Keep Job parameter defaults sensible — Tasks inherit them.
Use Sync Parameters after platform changes rather than recreating the Job, to preserve any Tasks already linked to it.
Use Execute Now to smoke-test a Job before scheduling Tasks against it.

Documentation

Job