I'd like to say it again please

I see a lot of posts on Twitter promoting the value of being skilled with data structure algorithms (DSA). I am going to state a rather unpopular opinion: outside of some unique situations, anything more than a familiarity with DSA is a waste. Therefore, all these interviews that insist you solve a DSA problem are also missing the point of an interview and therefore wasting a lot of time.

Please note, I am not saying DSA is useless. But in the day-to-day work for most engineers, it’s not relevant.

Example: You need to design a webpage that accepts input from an insurance holder to start a claim. You need a UI to collect the data, a function or two to validate it, and finally something to record the data—either by submitting it to a service or directly writing it to a database. In none of that work would in-depth knowledge of DSA help you.

So why is it such a hot topic? I think a large driver behind it is because it’s so important in interviewing. And it’s the interview format that really needs to change.

Interviewers should pick problem-solving tasks relevant to the day-to-day work the interviewee is expected to perform, not some theoretical college course material that, on a day-to-day basis, has no use. There are plenty of better ways to test problem-solving skills.

Take the example above. How can you turn that into interview question(s)? Using the example above: How would you do validation? How do you send data to a REST API? What’s the underlying protocol, mechanisms, etc.? How do you handle errors when communicating with the API?A DSA interview question will never prove an individual can think through the day-to-day problems encountered on the job. Of course, there are a few exceptions, but there’s way too much of this to be reasonable.