Student to SWE: What I did NOT expect

Last summer I took my first Software Engineering Internship, working at one of the nation’s largest chip-making companies (depending on the brand of laptop you’re using, you may see their name on the bottom right corner of the keyboard), and recently accepted their offer to transition to a full-time employee after graduation. There is plenty of literature and free advice available on transitioning from a student to a software engineer – some of which I read during the nervous anticipation period leading up to the start of my internship. And while much of that advice was good, I’d like to share a few tips that I didn’t see listed and would have appreciated knowing. I say this with the caveat that software engineering experiences do vary widely; this advice may or may not apply to your job. Make of it what you will.

1. Knowing lots of programming languages and having a wide array of technical skills is way less valuable than being able to problem-solve, read documentation, and learn new things.
2. Be prepared to spend less time actually programming and more time on any of the following: understanding someone else’s code, getting access to data and systems, navigating requirements, code reviews, research. See step 1.
3. Be ready to interact with new and unfamiliar systems. Ever used Azure Synapse? Snowflake? Tabular Editor? DAX Studio? SSIS Packages? There’s no way you could have prepared yourself for the sheer amount of new stuff – don’t even try. Read the documentation, ask questions, run tests, and don’t panic. Again, see step 1.
4. It is very rare that you will build a standalone system “from scratch”. A lot of our classwork involves a single .py file or an app built in two weeks with limited functionality. As it turns out, this is a lot easier than working with existing systems, existing code, and data sources that are a lot more complex than a single .csv or json file.
5. Think like a computer scientist. Remember in data structures class when your AVL tree just WOULD NOT WORK and you had to start over from scratch, internet advice wasn’t cutting it, and you eventually just wrote out pseudocode in your notebook until the algorithm finally made sense? Good. You’re going to need this – being able to work the problem makes all the difference. So, basically, see step 1.

You got this.