This looks to be a very interesting paper
https://lss.fnal.gov/archive/2023/pub/fermilab-pub-23-608-csaid-ppd.pdf
In this article we examine recent developments in the research area
concerning the creation of end-to-end models for the complete optimization
of measuring instruments. The models we consider rely on diferentiable
programming methods and on the specifcation of a software pipeline including
all factors impacting performance—from the data-generating processes to
their reconstruction and the extraction of inference on the parameters of
interest of a measuring instrument—along with the careful specifcation of a
utility function well aligned with the end goals of the experiment.
Building on previous studies originated within the MODE Collaboration,
we focus specifcally on applications involving instruments for particle physics
experimentation, as well as industrial and medical applications that share
the detection of radiation as their data-generating mechanism.