Workflows have the ability to execute tasks concurrently using the run_async field. This is ideal for tasks that are independent or rely on external systems (e.g., APIs, background jobs).

• To avoid blocking the pipeline while waiting on long-running tasks
• To run multiple tasks in parallel for performance
• To perform background actions (e.g., notifications, processing jobs)

You can use run_async on:

• ChipTask
• CodeTask
• DispatchTask
• TaskSet

When a task runs asynchronously, the result is not immediately usable - it is a task handle. You must store it using store_result_as or append_result_into and wait for it later.

- name: "Start background report"
  run_async: true
  chip: reports.generate
  params:
    user_id: user.id
  store_result_as: background_task

- name: "Wait for report"
  chip: reasonchip.async.wait_for
  params:
    task: background_task
  store_result_as: final_report
	

In this example:

• background_task holds the async task reference
• reasonchip.async.wait_for waits for it to finish
• The final result is stored in final_report

• Always use store_result_as with run_async
• To get the actual result, you must await the task for completion.
• Do not use the task handle directly - it is not the final result
• Ensure the task’s result is not needed immediately if run asynchronously
• Avoid using async with tasks that modify shared data unless properly synchronized