It literally prepares the audio for export, though. There is an export setting that allows you to choose the audio codec, channels, encoders, target true peak or LUFS level, and then normalize or "optimize" (built in limiting and compression) the audio to that level. For example, you can choose "Youtube" to optimize your audio automatically for Youtube. It also has a built-in dialogue leveler, for example. It has more audio capabilities built-in than many other video editors, although there are paid ones with the same capabilities.

You can easily do the same in Reaper, of course, but it will require you to export audio, then process externally, and import it again, so it's not the ideal workflow.

Oh so it hits the LUFS target automatically? This wasn't presented in the video I got sent. Great stuff then!