In many fields of research, there are some tasks that are self-contained and could be done effectively by someone who is not a full time member of a research group, but has the required skill and background. The task may be time-sensitive but often everyone is already fully tasked.
One might call them part-time or contract or temporary post-docs.
There are plenty of situations I can think of where this might not work, or would be a bad idea, but perhaps there are cases where the benefits outweigh any potential academic risks. They might involve a side-project, or an early foray into a potentially interesting area where there is no current expertise or time for someone to start from zero even though things might not pan out.
Because this would not be an employee situation, work would have to be done off-site. It might involve reading, writing, searching, programming, thinking, checking for errors (grammatical, logical, arithmetic...), or other tasks that could be exchanged electronically, and where the results could be checked and confirmed afterward. (You wouldn't want you primary data analysis done for example by "someone on the internet"!)
However I've never really heard of such a thing except in special circumstances where the individual has a previous relationship with the research group, and even then it can be difficult to work out the proper monetary mechanisms with the university administration, or funding agencies. "Consultant fees" are not a common line-item in research proposals.
Are there situations where this is changing? Are there any examples of micro-grants to get small jobs done in non-standard ways such as this? Is there even some better vocabulary to use to describe this kind of arrangement than "gig-egonomy" jobs?
edit: Seeing some the comments below it struck me that this might not be a very popular concept among regular post-docs. I don't think this type of resource will ever put normal post-doc positions at significant risk. But here I'm just asking if it happens, how might it be funded, and what it might be called, not if it's a good idea.