throwaway - I worked 10+ years at a bay area lab (out of college) without a MS/PhD but background in physics. Did software mostly. I left about a year ago to wo

throwaway - I worked 10+ years at a bay area lab (out of college) without a MS/PhD but background in physics. Did software mostly. I left about a year ago to work at a major bay area tech company. I’ve worked with people at FNAL, LBNL/NERSC, SLAC, JLab, LLNL, BNL, Argonne - as well as NCSA, CERN, NASA, INAF/IN2P3, tons of universities and NSF research facilities. I get paid now, but it’s because there was always chances to learn and challenges out in the national lab systems. my tech complaints are that the existing IT people in the national lab system really are old school. Most computing is done in batch farms/grid, and if your not that you have to be creative for your resources. Cloud isn’t really something that gets used. Build systems are ad-hoc. Lab overhead rarely pays for anything software for the purposes of software development. People are generally smart but really conservative to new tech. The NNSA programs are a bit more flexible than department of science programs (Sandia/Los Alamos might be most famous). On the office of science side there’s a lot more cross collaboration between labs. Simple things, like a $20/mo cloud service, are almost impossible to pay for. Expensive things, like a $400k purchase from Dell, is a piece of cake. Nobody is probably going to provide you a k8s cluster or anything, so running your own stuff/third party things is painful. Every national lab has a stupid opinion on which container tech to use and it’s never just docker or podman. Hope you like Singularity and Charlie Containers. That said, I still miss it. I will probably go back at some point. I don’t miss having to run so many things, fight central IT on small things, deal with purchase. I miss the people. Age distribution is bimodal, probably peaking at 27 and 55 would be my guess for many labs. Remote work is probably okay with Office of Science jobs, depending how close you might be to hardware. Probably not the case with NNSA work. It’s easy to write/collaborate with software as open source among all the California labs. LLNL is the most impressive there by a large margin. Berkeley software is under BSD usually. I was making 165 ish and now I make about 350k total compensation. Senior software would probably be around 145-185 at the very highest in bay area. If you can be an architect on a significant data intensive experiment with 200MM+ funding you might be able to get 200. Might get a 20k boost at LLNL if you are under NNSA related things.