How is quantum "measurement" actually done? There is a lot of talk in the fundamentals of QM about "measurement" and "observation", but never much specificity in what is going on at a basic level.
For example, an electron can be measured to have a certain position, at the expense of info about it's momentum, and vice versa. But what mechanisms are used to actually glean this information? What are we doing to the electron that tells us its position, and what is done differently to measure its momentum?
Presumably we have to fire some other particle at it, which becomes entangled with the electron and then contains the info, which through some chain of interaction eventually brings the info to a computer screen which we can read. But what are the kinds of measurement devices that experimental physicists actually use here?
Can these interactions be described with a Feynman diagram? Is there anything fundamentally different about a "measurement" interaction as opposed to any arbitrary particle interaction?
And further, isn't the "observation" part yet another normal particle interaction (inside a retina for example) with a specific path integral description? It feels to me like this talk of measurement and observation only serves to confuse the matter further.
 A: 
Can these interactions be described with a Feynman diagram?

Yes, of course.

Is there anything fundamentally different about a "measurement"
interaction as opposed to any arbitrary particle interaction?

No there is not.

But what mechanisms are used to actually glean this information?

Any interaction allowed by a Feynmann diagram. So, for example, moller scattering. But seeing the results of moller scattering doesn't tell you any of the particle's initial x/p with certainty - remember the cross section of a QFT process is itself a probability.
A: Historically speaking, quantum physics arrived when positivism in physics was in the ascendent. Hence both relativity and QM was phrased in that language and this accounts for the language of clocks and rods in the former and measurements and observers in the latter. Its worth noting that Ernst Mach, whose critique of Newtonian absolute time and space inspired the young Einstein as well as a whole generation of young German physicists, was deeply anti-metaphysical and strongly influenced the Vienna Circle of logical positivism.
It also stands to reason that when the very foundations of a subject look shaky, and which had once thought to be impregnable and impossible to think in any other way, then theorists are more likely to be careful with language and what they are signifying.
However, since then as people have become more comfortable with QM there has been a veritable effloressence of unobservable entities in physics. Beginning with virtual particles, continuing with quarks, and then strings, susy, D-branes, extra dimensions etc etc. An embarressment of riches and an embarressment of ontology. In fact, the entire quantum realm is unobservable: it is only that which is actualised through measurement that is observable. The existence of the quantum realm is inferred, not observed. This led Bohr to his relativisation of ontology (relative to a classical measuring appartus) and Grete Hermann's more radical departure where ontology is relativised to the outcomes of a measurement. A modern account of this is Rovelli's relational QM where he explicitly states that a measurement is nothing other than an interaction.
It's worth noting that this isn't the first relational ontology. Aristotle solution to Zeno's challenge to the question as to what consiltitutes change was in two parts. That change was primarily motion and motion was becoming which was potential, because continual motion is potential and being was actual and at rest (meaning not becoming and hence not undergoing change):

(263b3) So the reply we have to make to the question whether it is possible to traverse infinitely many parts (whether these are parts of time or distance) is that there is a sense in which it is possible and a sense in which they are not. If they exist actually, it is impossible; but if they exist potentially, it is possible.

Thus motion is a sequence of alterations between becoming which is potential, continuous and changeful and being which is actual, discrete and at rest (not undergoing change).
It is remarkable that Aristotle achieved a relative ontology of motion akin to that which we see in QM; but no more remarkable than how the Greek atomists prefigured the atomism of Dalton to Bohr. Except that the atomists theories fed directly into modern physics through Newton (his corpuscules) whilst Aristotles rival theory hasn't had the same impact. It might be worth adding here, that Aristotle saw his theory of change as a critique to that of the atomists and we see here in this, a similar critique that QM posed to modern day atomists. History repeats - but not exactly.
