Why Coding Came Easy (And What That Says About Math)

 Written with Claude

My coding journey started at 14. I had taken CS as an optional subject in school, and we learned to program in BASIC. For my final project, I tried to make a car racing game. I got stuck almost immediately: I couldn't figure out how to assign random obstacle coordinates to a variable and compare them to the car's coordinates to trigger a collision. I gave up and did Hangman instead.

That small defeat aside, something about coding felt natural to me in a way that math never did. I drifted mostly into biology after school but took a C++ class in college and a bioinformatics class in grad school, along with a course in computational cognitive neuroscience. Then coding became a professional necessity. A large part of modern biology, especially bioinformatics and large-scale data analysis, runs on code. I taught myself Python and R, and by the time I was doing postdocs in computational biology, both had become genuinely fluent languages for me. I was teaching others, not being taught.

Which brings me to a contrast I still don't fully understand. Math and coding are supposed to be close cousins, both precise, both logical, both unforgiving of vague thinking. And yet one of them took me a decade of struggle, embarrassing gaps, patient tutors, and YouTube binges to get even moderately good at. The other I picked up and ran with. I genuinely don't know why.

My best guess has to do with the nature of the abstraction involved. Both math and coding reward clear, rigorous thinking. But I think the kind of abstraction they ask for is qualitatively different. 

What's interesting is that CS theory seems to live firmly on the math side of this divide. When I encountered automata theory, the naturalness I felt with coding vanished completely. Finite state machines, formal languages, the abstract machinery of computation: I struggled with it in exactly the way I struggle with math. Which suggests to me that it wasn't coding I was good at so much as a particular kind of applied, concrete thinking that coding happens to reward. The abstraction ceiling was always there. I just didn't hit it until the theory showed up.