It is very difficult for a new graduate to know what a hiring manager is expecting of them. The lucky minority may have had the benefit of an internship doing something similar, but for those who have not, the gap between University and work is unknown, and it would not be unreasonable to feel like Christopher Columbus: confident that the Earth was round and that he would eventually hit land if he continued to sail west, but at the same time knowing nothing about where he was sailing.
If you are a manger hiring graduates for the first time, you might want to reflect on this and set your expectations appropriately.
Things University generally does pretty well:
- Ensuring that you know how to learn. This is, by far, the most important skill you can possibly have as it is the foundation of developing new skills.
- Ensuring that you have a good basic knowledge of everything covered in the scope of your degree. Good graduates generally know a bit about quite a few things.
- Ensuring that you know how to solve solved problems. Many graduates know how to code a balanced red-black tree from memory as well as knowing when (at least in theory) this would be a good data structure to choose. I have no clue about either of these things, but I have a good algorithms book and I know what I do not know (i.e. when I need to consult a suitable resource). Thus far, in over 25 years working as a software engineer, I have never had to code a balanced red-black tree, so this seems like a reasonable trade-off to me.
- Teaching you basic (as opposed to BASIC) programming skills.
- Teaching you algorithms, and good ways to handle data in general.
Things University generally doesn’t do so well:
- Teaching you about programming in the large – version control, code reviews, continuous integration, regression testing and the like. You will most likely be initially overwhelmed by the sheer size of some codebases, and I will not be surprised by this.
- Teaching you about control and management code. Most students are introduced to state machines and finite automata, so you have a good theoretical grounding, but control and management can be complex and messy – not something that is always amenable to academic study.
- Teaching real-world concurrency and parallel programming. There are exceptions (functional programming courses often teach this as it is an area where FP has significant advantages over imperative programming), but I have interviewed relatively few students who have much understanding of the basic synchronisation primitives.
- Debugging. Most of the projects students are expected to undertake are not sufficiently complex to allow you to develop good debugging techniques.
Most of my expectations should now be pretty clear: I expect you to have a good understanding in the areas Universities generally do quite well, and maybe a very basic understanding in some of the areas that are done somewhat less well.
There are a few other basics that you should have picked up long before you started your tertiary studies.
- You need to be able to communicate in clear and reasonably concise English – both written and spoken.
- You need to be able to get on with a variety of people of different backgrounds, ages and opinions in a professional environment. You don’t need to be super-outgoing, or be everyone’s best mate, but you are going to be spending over 1600 hours with your close team mates every year, so getting on at a basic level is important.