| Zkuq said: As far as I know, game programmers are highly valued, at least in some fields. I suspect it's especially people close to the engine, where the work is more technical in a way (performance requirements, math etc.), but I don't know for sure. My point is that at least some programmers in the gaming industry probably actually are much better developers than you might think at first. |
Right. These days, those are the 1 in 100 (i.e Epic's well-paid SWEs) who are game engine engineers and maybe some indie developers who are writing low-level (or near it) code on the CPU/GPU.
Although that doesn't take away from the complexity of UI engineering, scripting, and other aspects of games either. Just that the best of these areas can be better paid in other industries and likely flee to those other industries as they gain experience and want to make more money.
| IcaroRibeiro said: I can totally see why gaming engineering would be harder than data science. I took some gaming development classes and found it to be quite challenging even when using engines like Unity that streamline a lot of the work. I can also see why middleware engineers would make more money. It requires really specific skill set: Physics, math, graphical programming, low-level coding, etc DS is only hard (in technical aspects) when working in specific problems, mostly academic/research subjects. Most industry problems will not require advanced statistical modeling Data jobs are high paid imo because a) It's a new and growing field that take a while to create even entry-level workers (unlike software engineering which is more hands-on) and b) Computer Science majors struggle with statistics and math, while math/statistics majors struggle with software engineering, so someone with both skills tend to be highly valued from what I see |
Yeah the hardest parts of the data science role, in my opinion, is the soft-skills and business-knowledge DS require to be successful. Though I do think there are some technical challenges that arise often in the role too, that just aren't as apparent because people (who aren't data scientists) don't quite know what makes a DS "good" or "bad" at the job. You can be a mediocre data scientist who just tries to throw well-performing models at business problems that the models aren't well suited for and doesn't really consider out-of-distribution contexts, and for which interpretability of the model is a nightmare to the point of not adding value to the stakeholders or a decent data-scientist who knows when and where to use certain tools and how to report the results in an understandable manner.
Data Science is just a lot more open-ended of a role than most non-analytical Software Engineering.
