In college, my major was math and biochem (with an eye toward computational biology or computational chemistry). So no, I haven't always been a developer, my interests were very wide: everything from systems biology to quantum mechanics. So wide that I ultimately dropped out at the start of my senior year, rather than choose some specialty for graduate school. I don't regret that, since at least I have a couple friends who did go on to different grad programs (physical chemistry, molecular biology) and I still appreciate the independence and freedom that comes with dropping out.

Anyway, I always wanted to be more of a big picture / theoretical guy and sort of poo-poohed programming. I thought it was more important to have deep understanding of mathematical formulas and creative theoretical research ideas (the passion). But I've come to appreciate that implementing such formulas/ideas as code (the skills) is much harder than I imagined (as is any type of coding), and also the best way to gain a deep understanding.

So that's how I came back to programming in the past few years, which I always treated rather casually growing up (started with some visual basic in my early teens).

How did you end up in banking support work? Maybe thought it was an avenue to wall street quant/HFT work (wild guess)? In Cal Newport's book he does say there are certain jobs where the best option is to leave, eg when there's little opportunity for growth and learning (building "career capital"). But his overall point is to stop chasing a passion by looking for an expected dream job right out of college. Instead, develop it over time by earning and leveraging/parlaying career capital. Going through the motions, gaining experience, and developing valuable skills through hard work. You should check it out (I finally found a pdf online only the other day), sounds like it will be worth your read.