This is not a very original recommendation. In fact a major selling point of String Theory is that one can manage to derive both Einstein's Field equations and

This is not a very original recommendation. In fact a major selling point of String Theory is that one can manage to derive both Einstein's Field equations and quantum field theory scattering amplitudes from its equations. This approach is currently the only one that can claim that for itself. Of course people like Hossenfelder never made an effort to understand String Theory in detail, so they can only make first order observations about the current state of the field. It is also not true that no progress has been made in the understanding of String Theory in the last 40 years and it still seems like the best bet that could eventually generate a fundamental theory. What is missing is still a lot though: - We don't seem to possess the correct mathematics to develop a non-perturbative formulation of String Theory and there are too many potential string backgrounds that we could expand around. - It is also hard to derive the matter content of low energy effective actions from most brane configurations. String theory provided major insight into non-perturbative quantum field theory as well. There is tons of examples, let me highlight one of them: The discovery of the Amplituhedron (Arkani-Hamed et al., 2012) was preceded by the discovery of the BCFW recursion relation (Britto et al., 2005), which in turn was motivated a relationship between perturbative Yang-Mills theory and the instanton expansion of a certain string theory in twistor space (Witten, 2003).