## How Long Do You Have to Solve Technical Questions?

### Posted on Sunday, November 13th, 2011

Before telling you how long you have to solve technical questions, let me ask you a question.

Your math teacher assigns you a problem. It takes you 5 minutes to solve it. Is that quick or slow?

Hopefully, you’re looking at this question confused. 5 minutes might be unacceptably long to compute, say, the square root of 16, but extremely quick to solve a more complex proof. I didn’t tell you what the question is, so how can you tell me if 5 minutes is quick or slow?

So how long do you have to solve technical questions?

As in the above situation, it totally depends on the complexity of the question. A simple factual question might take seconds. A reasonably straightforward algorithm question might take a couple minutes. But a complex algorithm could take 30 minutes or more to solve.

That said, in a technical typical interview, a candidate typically solves one or two coding / algorithm questions. That’s an average though, over typical questions; be very careful about thinking, “gee, I solved three questions, I must have done great!” Solving ten questions does not mean that you’ve done well, if they were easy questions. Likewise, you might have done extremely well without even finishing one question.

So how do you know how well you’ve done?

You don’t (really, really, I promise you, you don’t!). Your interview performance is impossible to judge.

Categorized as Uncategorized

### 3 Responses to“How Long Do You Have to Solve Technical Questions?”

• #### Abhishek Dilliwal says:November 14th, 2011 at 6:33 am

rightly said, also what if the interviewer is expecting a better approach rather than just correct answer?

• #### Sanjeev Kulkarni says:November 14th, 2011 at 7:37 am

A tiny bit of great reinforcement of Laws of Interview!

• #### Gayle Laakmann says:November 14th, 2011 at 5:29 pm

@Abhishek It is (almost) always about more than getting the right answer. First of all, most questions are not judged on a correct / incorrect basis. Many algorithms might work but be suboptimal. Secondly, the speed to get the solution, the approach taken, the mistakes made, the cleanness of the code, etc, all matter.

It’s very rare for an interview to be looking for merely the “correct answer” (whatever that means).