Yes, he is. (At least I hope he is.
). I’m not so sure about sidechaning the master bus compressor, though. That, kind-a, defeats the purpose. Of course, you could use a multiband compressor. That is a compressor that deals with various frequency bands separately. However, now we’re moving into the territory of mastering, which is an art in itself.
It’s sound to me that you are not only new to Cubase but new to recording, period. This is a huge topic. Far to big to be covered here. I suggest that you invest in a course, such as Ask Videos Audio Concepts series.
https://www.askvideo.com/courses/application/audioconcepts
That way I’m sure you’d be in a much better position to figure out what you’re doing wrong, and avoid it in the first place.
As for your level problem. If you by “when the drum kick in”, you mean when the drums enter at 0:50, I can neither hear or see any problem. The level of the flute sounds the same. There may be a perceived lowering in the level of the guitar, but that’s only natural as it is being partially masked by the drums. Frequency wise.
However, I do notice another problem. You have mixed it a bit too hot! Starting at 1:40 there are some clipped waveforms every now and than. These may not only cause audible distortion but, more seriously, may interfere with mp3 encoders. It’s highly recommended that the level never reaches 0dB (as it does in several places in your track). You can solve this by inserting a brickwall limiter on your main output buss, set to at least -0.5 db. Also make sure that it’s only triggered occasionally.
I trust that you’re recording at 24 or 32-bit resolution. Then you can easily mix with the brickwall limiter set to -3dB. The nosefloor will still be far below human hearing level. Always mix with full dynamic level. You can always add compression at the mastering pass. It’s more or less impossible to remove it, though.