But ultimately I would like to see something like this as a subclass of an Artificer/Alchemist class. Although as a test run for how prestige classes might work.... it isn't bad, but this looks a better fit for where I think they may end up with their Artificer/Alchemist why not simply make it part of that?
It's first and foremost a matter of
access rights, in the sense of the question: "should this archetype be accessible to only one class or more?".
In the case of this Runecaster, it's more than clear that they wanted it to be
not a specialized Wizard, but instead something separate from each class' own source of spellcasting power. Actually, they wanted it to be even separate from having spellcasting powers in the first place! They just decided that
everybody should be able to learn how to use runes, and that's why it is not a subclass.
It could have of course been a feats chain too.
I've been thinking about a few of the problems with Prestige classes -
Runes and Things - and I'm really, really dubious about the concept in 5E. I'm not utterly opposed to it, because I don't like shutting down avenues of design. There are wonderful things that can appear if you allow them to...
...but also a lot of horrid, unbalanced drek. Hmm...
I totally agree with your fear that Prestige Classes are very hard to design, in fact I was about to post the same here yesterday, then decided to just give it more thoughts...
Let's compare the design challenges of feat chains VS subclasses VS prestige classes, assuming you want to implement basically the same thing (i.e. a few consecutive special abilities):
1)
Feat chains: you can always choose to specify a minimum character level requirement for a feat if you want, so you can design those abilities freely as you wish, and then just ask yourself "what should be the minimum level of ability X and Y", then slap such level to the feat as a requirement.
The only downside is that some classes might get feats at a significantly higher level than required: e.g. you design a feat with 5th-level requirement, then maybe some classes get a feat at 4th level (too early) and the next at 9th, so they have to wait 4 more levels. But it's not a huge downside.
2)
Subclasses: the levels are fixed, so you rather first take a look at what those levels are, and then design/tweak the special abilities to match the level.
No downside other the fact that you are not completely free, and some classes have remarkably few subclass levels with a huge gap.
3)
Prestige classes: you can set a requirement for the first level, then the following levels are consecutive. So you have to design all special abilities to be appropriate for the level range. You can shift everything up or down in level, but that's it. If you decide that 6th level is where the Prestige Class starts, and you have 5 levels worth of special abilities, all of them must be appropriate for levels 6th-10th.
The downside is that you are even more restricted in terms of design, and furthermore the whole class might even end up significantly less attractive if you take it later than as-soon-as-possible, if the special abilities don't scale up well. [One unprecedented design possibility would be to have
each level of the prestige class to have its own requirement, also in terms of minimum level].
Furthermore, you are at least required to decide the HD of this class (tho at least compared to 3e you don't have to decide BAB, ST, skill points etc.) which is extra work beyond the point of those special abilities, which are really the purpose of the prestige class. The Runecaster is already an example of this small additional problem, with its d8 HD being clearly a compromise value which tries not to let non-Wizards lose too much. If the Runecaster was a feat chain or a subclass, there would be no such problem.