I resonate with this. I’m not a SWE, but still run the risk of spending too much time in theory and not vetting my work in the real world.
There is totally room for software in the real world, even in the sort of job where you take a shower at the end of the day. Perhaps an “applied sciences” approach is best. It isn’t that we don’t do the “computery stuff” (as my wife would say) but rather it is always best to focus on how that impacts the real world. Are we optimizing fertilizer and irrigation application to boost real crops for less input cost? Are we maximizing the customer experience for a painful process to make it “spooky easy” compared to the other ways of buying things we need.
Gum shoe software still has a chance to have a huge impact. There’s a chance that your skills or your field are not broken but perhaps the company is.