RDM thoughts and tuning needed

[Archwizard]

I agree that our AoE feels far too slow. It's not an issue that it's lower than our ST buildup relatively, considering Moulinet is still cheaper than any other melee skill and we don't need to worry about an AoE combo... but Scatter/Impact in particular returns less Mana than either of the other AoEs (which themselves return more than Jolt/II). I might recommend Scatter/Impact be buffed to 4/4 Mana per cast, or perhaps reduce Veraero/thunder II by 1 Mana each and change Scatter/Impact to 2/2 Mana per target (maybe with a cap) -- the latter of which would go a long way towards a permanent address to the issue (more akin to the Enhanced Scatter of old) without just cloning our single-target gains.

At the same time, I think the pre-Impact portions of our AoE (Scatter, Moulinet, Ver- II spells pre-78) could be uptuned just a little, particularly Moulinet since it's weaker and less satisfying than Impact despite being used about a third as often. While I acknowledge that Moulinet can be used multiple times in a row and has a shorter GCD, those advantages are reliant upon you having already pooled the Mana to use it multiple times in a row and hitting every GCD. I might even agree to a small potency nerf to Impact if it was made up for by a stronger portion on Moulinet (and perhaps a smaller gap with the Ver- II spells post-78).
I don't mind it being a cone, however. You can dash to the nearest enemy, hack at the crowd, then jump back without putting yourself deep enough into the killbox to be at risk of taking a hit meant for the tank. Plus it just feels more thematically "melee" this way.

As for Engagement, I agree that it feels like a waste of a slot right now -- while I grant that it is useful in niche situations when Displacement is not a viable option, I've had fewer instances overall in Shadowbringers where that was a concern; I can count on one hand the number of times I've had to spend the CD on Engagement since I got it, especially since every boss fight I've encountered so far has either had ample room on the platform or straight-up barriers to back into. Any time I have used it was to clear the CD during a mechanic before I hit Manafication.
The way I see it, there are three options to avoid Engagement simply being a bloat skill:
1) Replace Engagement, and remove the damage from Corps-a-corps and Displacement so that we can use them purely as optional movement tools.
2) Remove Engagement but keep the damage on other skills, and combine Displacement into one action with Corps-a-corps, where casting within a short period (say, 10-15 sec, enough time to combo) causes you to jump back to the location you dashed in from, so you can better control where you'll land.
3) Keep both the damage and Engagement, but make it a niche sidegrade over Displacement (other than removing the backstep, which is a net neutral) to justify its existence as a separate button beyond being a backup Displacement, such as AoE damage.

I agree on Embolden. It seems to me like Embolden was the devs' attempt to make a pseudo En-elemental buff for the party (particularly if we're synching it with our melee combo), but there were other ways to preserve that flavor. Even when it was introduced, there were more buffs that affected melee damage than there were for our fellow casters, and despite the removal of most of those skills, we uniquely persist in a manner that punishes pairing us with other casters. At this point I would much rather have had a group Haste buff, or perhaps a "add X% unaspected magic damage to all attacks" skill other casters could get into.

Current numbers show that since its nerf to 220 potency, Reprise is not worth using over simply overcapping our Mana, and community consensus is that it's too weak to justify spending 10/10 Mana on what is effectively our Scathe. Napkin math from theorycrafters I've consulted shows it needs a minimum of around ~260 potency to justify using it in an overcap situation or to delay our melee combo, and having it be a 300 potency skill for the cost of 2-3 generators was a good sweet spot, which compares well to 3-target Moulinet.

On a somewhat unrelated note from the OP, is it just me or does it feel like our MP is being burned through more quickly, even without Verraise? I still feel like I'm playing CD-to-CD with Lucid Dreaming (despite the supposed "net gain" in potency)...