Most of developers working in "finance" don't know or care about trading, and the few ones who aspire at this... there's no better representation of their context than this:
Nowadays the two professions are becoming almost inseparatable, no? Traders must understand technology and developers, in the industry, must understand trading. With all due respect to real traders, I'm sure there are plenty of developers who would consider trading as a step down.
Writing software and trading are two entirely different endeavors. Switching between the two could be difficult.....or a steep learning curve.
Seems to me you'd need to first understand trading before then translating that understanding into code. IOW... if you don't understand the trade(s), just what are you coding for??
Absolutely !!! Where both are difficult to learn in their own rights, trading for me was much harder to learn than learning to program, and now it is easier for me to tell a programmer what I require than if I was just a programmer. I often find if you were a programmer first, much tougher to learn trading as trading is a non absolute, often time patterns happen because of more emotional states of people, how they view currently and if underfunded.
I think being a software engineer can only help. I agree that it's best to learn how to trade first, but I can see a few advantages that a software engineer would have over traders without software experience: 1. They can develop their own charting platform. They fully control how data is processed from the tick level. This allows them to do things such as merge multiple datafeeds and have multiple layers of processing that would be a lot more difficult if not impossible to do in a standard charting package that's really only intended to show time-based bars with moving-average based indicators. 2. Easier integration of 3rd party data mining and machine learning software. 3. Easier to perform ad-hoc backtesting. 4. Automation. I have a scanner that I run nightly to identify good candidates to add to my watch list. I simply prefer reviewing a few charts vs. 500-1000 every night.
>> They can develop their own charting platform. They fully control how data is processed from the tick level. This allows them to do things such as merge multiple datafeeds and have multiple layers of processing that would be a lot more difficult if not impossible to do in a standard charting Tommy Reid was a trader. Actually he was a developer, but in his spare time he also did some trading and because of that he thought of himself as a trader. He did software development somehow by obligation, albeit he liked this one too, but his true vocation was trading, he considered. In spite of that he was doing serious trading extremely seldom, about once a year. Otherwise he was extremely busy with software development tasks. But once a year he let everything aside and started trading. Ne never managed to get anything (profitable) and neither did he have proper trading tools (models, backtests, quality marketdata, fast executions, automated systematic engines) but the trader, with or without tools, is still a trader. And if anyone happened to ask Tommy what was his job, he smiled contemptuously and didn't say nothing, because he considered you can see on his face he's a trader and if some guy doesn't get this thing, then it's that guy's business, because Tommy is still a trader.
I had to learn over 3 years to program well, and while learning I continued to trade, some point wanted more and hired a staff to do all the programming of what you did and more was much easier than your route to me. But we each like different areas of trading which is what this still floats my boat after 39 years. I prefer more studying than anything doing my week. But there are many who know how to program and fail at trading, at least they can fall back to working for others or themselves if they can sell themselves, whereas a trader can fail at programming and continue to work for himself and just easier to hire others. Never run out of ideas to backtest.