Can you give more information about your setup? Where is the Camera2D in the node tree (if it a child of player or loose in the scene). What problems are you facing?

For this kind of effect (like Zelda or Celeste) it is probably best to have the camera loose at a higher level in the tree. It won't move when you move the character (but can if you code it specifically). Then when you get near to the edge of the screen, you can detect the player is out of bounds (by checking the player position against the current screen bounds in world space). Then check the direction to see which should be the next screen.

If the screens are all the same size like Zelda, the math is easier, because you can just do some translation to get the screen position/bounds. If it's like Celeste, where some screens have different sizes, you probably want to create an Area2D on top of each screen (sized to the full screen dimensions) and you can use that for your calculations.

Then finally you just take the new screen position and tween the camera to that. Sorry if that explanations if vague, maybe it will help you get some ideas.

My Camera Node is a child of the KinematicBody2D (the player)

Also I'm gonna make the screens atleast in increments of the screen size 320x240 like making 2 screens long then transition

You could try animating the Camera2D bounds. I haven't tried that but maybe it would work. Like using "linear_interpolate" on the Camera2D x bounds.

3 months later
a year later