Non-academic careers and physics PhDs

With so many large-scale events happening right now (the pandemic, resulting economic displacement, the awful killing of George Floyd and resulting protests and unrest, federal moves regarding international students), it's hard not to feel like blogging is a comparatively self-indulgent activity.  Still, it is a way to try to restore a feeling of normalcy.  

The Pizza Perusing Physicist had asked, in this comment, if I could offer any guidance about non-academic careers for physics PhDs (including specific fields and career paths), beyond cliches about how PhD skills are valued by many employers.  I don't have any enormous font of wisdom on which to draw, but I do have a few points:

• I do strongly recommend reading A PhD is Not Enough.  It's a bit older now, but has good insights.
• It is interesting to look at statistics on where people actually land.  According to the AIP, about a half of physics PhDs take initial academic jobs (postdocs and others); a third go to the private sector; and 14% go to government positions.  Similarly, you can see the skills that recent PhDs say they use in their jobs.  
• I found it particularly interesting to read the comments from people ten years out from their degrees, since they have some greater perspective - seriously, check out that document.
• Those latter two AIP documents show why "PhD skills are valued by employers" has become cliched - it's true.
• In terms of non-academic career options for physics PhDs, there really are a wide variety, though like any career trajectory a great deal depends on the skills, flexibility, and foresight of the person.  Technical problem solving is a skill that a PhD should have learned - how to break big problems up into smaller ones, how to consider alternatives and come up with ways to test those, etc.  There is a often a blurry line between physics and some types of engineering, and it is not uncommon for physics doctorates to get jobs at companies that design and manufacture stuff - as a condensed matter person, I have known people who have gone to work at places like Intel, Motorola, Seagate, National Instruments, Keysight (formerly Agilent), Northrup Grumman, Lockheed, Boeing, etc.  It is true that it can be hard to get your foot in the door and even know what options are available.  I wish I had some silver bullet on this, but your best bets are research (into job openings), networking, and career fairs including at professional conferences.  Startups are also a possibility, though those come with their own risks.  Bear in mind that your detailed technical knowledge might not be what companies are looking for - I have seen experimentalist doctoral students go be very successful doing large-scale data analysis for oil services firms, for example.  Likewise, many people in the bioengineering and medical instrumentation fields have physics backgrounds.
• If academia isn't for you, start looking around on the side early on.  Get an idea of the choices and a feel for what interests you. 
• Make sure you're acquiring skills as well as getting your research done.  Learning how to program, how to manipulate and analyze large data sets, statistical methods - these are generally useful, even if the specific techniques evolve rapidly.
• Communication at all levels is a skill - work at it.  Get practice writing, from very short documents (summarize your research in 150 words so that a non-expert can get a sense of it) to papers to the thesis.  Being able to write and explain yourself is essential in any high level career.  Get practice speaking with comfort, from presentations to more informal 1-on-1 interactions.  Stage presence is a skill, meaning it can be learned.  
• Don't discount think tanks/analysis firms/patent firms - people who can tell the difference between reality and creative marketing language (whether about products or policies) are greatly valued.
• Similarly, don't discount public policy or public service.  The fraction of technically skilled people in elected office in the US is woefully small (while the chancellor of Germany has a PhD in quantum chemistry).  These days, governing and policy making would absolutely benefit from an infusion of people who actually know what technology is and how it works, and can tell the difference between an actual study and a press release.

I'm sure more things will occur to me after I publish this.  There is no one-size-fits-all answer, but that's probably a good thing.