LLMs make it trivially easy to add parallel processing boilerplate to one-off scripts that would otherwise run sequentially.