Strategy to solve lot of leetcode questions fast [closed]

Question

I am new to Python , but have to learn it for interviews. I have experience in shell scripting but no FAANG accepts coding interviews in shell.

So every question I have to go over each of the algoexpert conceptual videos (takes 45- 1 hour to understand) and then write them down on paper to gain confidence, later on the laptop without looking. But it is taking a good 2 hours for every question.

Not sure how am I going to solve a lot of LC qns like others doing with interviews scheduled in a few weeks' time. I already bombed one hackerrank test due to lack of this experience.

I have a full-time role and a lot of visibility at work, so it doesn't allow me to coast. YOE:14

Any other strategy you guys can suggest to get LC expertise faster than my current rate?

Answer 1

I assume you've already decided you're aiming for a company using these sorts of problems, you have a limited amount of time to practice, and you want to make the most of it. A no-nonsense, practical answer: single best thing you can do is get Leetcode premium subscription.

This solves three problems for you:

1. You have access to reasonably well curated answers, often with alternative solutions, right next to the problems you solve. This saves you time on doing your own research and is definitely more efficient than watching videos. Go back to those for problems you can't wrap your head around and really need to put some extra effort into,
2. You have access to Top 100 questions lists for different companies. Following those is a good way to keep yourself looking at high frequency, common questions, so you'll be covering useful material rather than going deep into some off-tangent topic,
3. You have access to mock interview sets, which will choose problems to solve for you. That's important since you will inevitably develop some sort of bias in what problems you look at, either picking things you know how to solve, or picking things you're most concerned about and neglecting other kinds of problems.

After that it's just about putting in the time and effort to practice. Don't wait with mock interviews too long as they tend to uncover blind spots quite effectively. Furthermore, try to write down the complete solution, don't depend on write-compile-fix cycle too much - many companies use whiteboards or scratchpads that don't come with a compiler, and you'll need to be comfortable in solving these problems without that support.

Answer 2

The leetcode test isn't the final verdict. They aren't supposed to be challenging Rather, it's an easy way to filter out the totally imcompetent candidates, and if you pass it, the real reviews can begin.

1. Learn Python. Take the Python tutorial, read through it and play around with the examples until you're confident that you understand how they work. For just interviewing you don't need to know libraries, because you'll be implementing the exercises on your own. For actual work knowing the standard library and most common third party libraries and knowing how to search for libraries is important. But so is being able to implement your own if the thing you need isn't available.

2. Learn computer science. Take any intro level university algorithms textbook or online course, and do the exercises in Python. This is the crucial part: you need to be able to apply what you just read, and this needs practice. As you already noticed, when compared to text videos are a huge waste of time.