Understanding the stock market before coding the app, and before investing

I've had an investing app idea in mind for a while. I haven't built it yet. Not because the subject is too technical, but because I've set myself a rule: really understand the stock market before writing any code around it, and before putting one more euro into it myself than I actually understand.

The competent developer's trap

The risk, as a developer, is believing that knowing how to build a tool is the same as understanding the domain it represents. I can write a compound return simulator in an hour. That doesn't mean I understand risk psychology, real-world taxation, or why most individual investors underperform the very market they're trying to beat.

It's the same trap I criticize when it comes to generative AI: accepting an answer that looks correct without having verified that it actually is. Coding a financial tool without understanding its fundamentals produces a tool that looks serious and can actively mislead.

What this actually means in practice

Before touching any code, I spend time reading, simulating by hand, understanding why certain investment decisions that seem obvious actually aren't (market timing, real versus perceived diversification, the cost of compounded fees over twenty years). Only once that understanding is solid does the question of the software product become relevant.

The parallel with my job

This discipline isn't different from the one I apply to counterparty risk: understand the mathematical model before coding an implementation of it, rather than trusting a formula you copy without challenging it. The domain changes, the method stays the same.