There's a nice dictum from Hamming that goes something like, "In engineering, if you don't know what you're doing, you shouldn't be doing it. In science, if you

There's a nice dictum from Hamming that goes something like, "In engineering, if you don't know what you're doing, you shouldn't be doing it. In science, if you know what you're doing, you shouldn't be doing it." You got a degree in computer science, but you are employed as an engineer. In most projects you should have at most zero computer science challenges. Once in a while you find a project where a quick back of the envelope calculation says that spending a certain amount of time up front trying to solve a particular research question is worth doing. That's unusual. The whole point of having a mature discipline of computer science is so that we can operate as engineers. The interesting problems being wrapped in ready-made libraries and SaaS services is an engineering success. As a profession we have managed to take a hard thing and change the cost of it from some number of expensive engineers for some years plus maintenance on it to $29.99/month. But that means that if you want to work on interesting problems of this kind, you need to be in a company that is building that switch from requiring an expensive team to off the shelf. And to identify that, you need to have a broad understanding of your domain, and where the pain points are that are shared but aren't worth the money for individual companies to solve. Or you need to be at a company that is big enough that it can afford to work on those problems. Boeing, for example, has dumped a huge amount of effort into spline libraries beyond what the rest of the world has. At their scale and for what they do, it was worth the investment. Facebook wrote a replacement database backend (RocksDB) because the space savings it gave them over InnoDB far outweighed the cost of having a team build it. It's a well known fact in other engineering disciplines that all the detailed physics and math you learn in college isn't what you use most of the time. The Engineering Commons podcast has [links](http://theengineeringcommons.com/episode-124-mad-skillz/ http://theengineeringcommons.com/episode-124-mad-skillz/) in one of its episodes to a number of studies MIT did on this subject. Your other option is to say, "I don't want to be an engineer. I want to be a scientist." And then you have to look into the rather fraught career path of a researcher.