Teach Yourself Computer Science

I got an interview that involved part 2 (substring search), which I solved using KMP.

I like to compare CS to medicine.

Most topics of it aren't difficult to understand, but overall it's just a lot of information. If you want to get a comprehensive CS-education, you have to accept that it takes years of studying, learning and practise.

I would like to add this related question to the discussion: what resources/courses/learning techniques do you and others like you who have had an undergraduate Computer Science degree for a few years and are interested in academia but not sure what steps to take next use to continue your study? At some point I would also like to return for a PhD or Master's degree and I would like to keep myself mentally trained, but there are so few resources for higher-level study. All of the online resources I've tried are almost exclusively introductory and don't provide the challenge I crave.

Does anyone taken higher-level courses at a local university in their free time just for the experience and to be surrounded by like-minded individuals while having a professional career?

The FAANG interview process is pretty standardized, if you're pressed on time I suggest you learn and master just 2 things: data structures and system design.

Are you thinking or debating if getting a deeper understanding of CS will help you or your career?

I am thinking of the same. Even though I have a Bachelor of Science (BSc) in Computer Science, it has been 14 years that I actually used any of it. I regret it sometimes since I know what it is to know how things work underneath the high level languages (PHP, JavaScript and Python for me).

Lately I have started taking more interest in CS and I think part of the reason is Rust. You could be learning Go, Swift or any similar language and come to similar feelings.

In Python (my language of choice for the last 7 years along with JavaScript) I do not really have to care about pointers, stack or heap allocation, how memory gets create or cleaned, etc. I also do not need to know if there is a virtual machine, a garbage collector, etc. Data types and how they are handled are somewhat important but I rarely get slapped by the Python interpreter for mistakes.

To me it seems that a lot of developers are choosing to go into more typed and lower level (closer to metal) languages. My gut feeling says this is happening because we can not crank up core speed from 2GHz to 3Ghz to 4Ghz and beyond as easily as we did pre-2Ghz era. Yes we hit ceilings of per-core power. But we have more users than ever before and the world seems to be eager to bring computing to even larger audiences.

All this is great, but it means developers need more refined and powerful set of tools that can work in low-power environments, boot up fast and deliver more computing throughput. The high levels of abstractions that most of us have enjoyed will stay but there will be new territories where you want to push code that is way, way more efficient. If we want to work with them, we need to know not only how to push a Template into an `` but about the metal that is underneath. From how browsers work internally to how networks work, how processes work or communicate, how memory is allocated and so forth.

So this is a choice we have to make individually. I choose to learn and I am going with Rust and spending time to learn CS again. It is a slow process though, I have full time commitment to my product as a solo founder. Cheers!