Yes, the automation should be designed to use commands that use windows drivers instead of mocking keyboard/ mouse actions to allow parallel actions. We also have Picture in Picture (PiP) as an experimental feature in 2020.4.
We have the public StudioX FAQ, so might as well make it easily discoverable as it will likely always cover more StudioX specific questions than a single AREP FAQ
Yes, the automations can be built in such a way that it should use commands from windows drivers directly instead of keyboard/ mouse actions to allow another task to execute.
In case of active UI elements interaction, we also have Picture in Picture (PiP) mode where automation runs in a small window over the desktop.
Hope this was helpful, feel free to drop comments.