I feel you; I too went through the same sort of thing -- PhD in CS from Cambridge, girlfriend in London who thought the center of the universe was London and pa

I feel you; I too went through the same sort of thing -- PhD in CS from Cambridge, girlfriend in London who thought the center of the universe was London and parents back home screaming at me to 'get a house and settle down in London'. I was slogging it, being paid incredibly well only to end up at work before the sun came up, back home when the sun had gone down, forever damp, wet, cold, windy even when it was sunny outside. A lot of your post centers around your living situation and being unable to buy a house in London even though you're well paid. You don't like London and hate it -- don't force yourself to live in it. I did that, and I got really really sick. The stress basically broke me. Like you I was scared -- scared that I'm unemployable and tied into my job, scared that I have no skillset, scared that I'm not good enough. I ended up being pushed, and that was not great. In the end I realized all the things I thought would mean the end of my life didn't mean anything at all: - Not living in London didn't matter - Not owning a house in London didn't matter - Not making my parents didn't matter - Not being with my ex-girlfriend didn't matter. What did matter was being happy and doing things that made me happy. I spent a year sitting at home learning about digital design because it seemed interesting. Today, I'm in a research commercial research position, doing well and happier than I've ever been. The (long winded) point I'm making is -- sometimes deep down (or not so deep down) we KNOW something isn't for us but we plug away anyway. Once you know, you know. Start making steps to change, do not continue to fight it. If you have to fight every day it's not for you. Don't be like me and break into two. Just start taking steps to fix it. Don't be afraid if the steps mean some uncertainty. Having said that, here are some practical ideas: - Have you thought about a research position? Take a look at the Alan Turing institute, they usually have interesting positions available. - Take a look at DAMTP/Stats @ Cambridge. They are really getting into ML/Stats for machine learning. Imperial too is ramping up their research teams in that area. - Have you thought about applying to one of the bigger, older insurance companies? Everything I'm hearing leads me to believe they want to ramp up the ML/technical side of the underwriting and claims process but don't have the technology skills to pursue it. If any of these seem interesting, ping me. I can make intros.