This video is free to watch - hope you enjoy it! If you like what you see, there's a ton more where that came from...

Buy Now

Databases: How Would You Implement...?

Screener questions tend to have a bit more broad and, typically, based on your experience... sometimes not. Think about how you would answer this one about databases.

This is premium content - you can own it in just 30 seconds!

Buy Now

You’re sitting at your kitchen table at home and the house is quiet and hopefully remains so. You have a glass of water next to you and you’re looking forward to the initial interview, even if you’re a bit nervous. You make sure you’re not backlit by sitting in front of a bright window and you check your audio/video settings for the third time - you don’t want to waste time with “can you hear me???”

The clock ticks 1pm and you click the link for the Zoom meeting and see a person is waiting for you there. Her name is Samantha and she’s the head of the data team at Company X.

Good morning, how are you today! Thank you so much for taking the time to talk with us. Before I go on - are there any questions you have for me?

You know it’s good form to ask at least a few questions and you do, inquiring about the job and maybe a bit more about Samantha’s duties. Eight minutes go by and she moves the conversation back toward you.

I looked over your resume and I can see you’ve done a lot of work with databases. Tell me about…

She asks you a few light questions about a job you had as the “developer DBA” at your last startup using MongoDB and, more recently, about some light analytics work you did using PostgreSQL. She pauses… here it comes…

The Question

The smile fades a bit and her face becomes a bit more serious - a queue for you so you know a technical question is about to be asked…

If I was to ask you to implement an indexing engine for a database we were developing, how do you think you would implement it? Just a simple one at first…

You take a second and think about the question. You remember that it’s a good idea to make sure you understand what’s being asked, so you begin to form some questions in your mind:

  • I assume this is a fictional product and that it has no indexing just yet? Yes, that’s correct.
  • Will the index be used to for string queries or more structured data? String data.
  • I’m also going to assume this index will be used for rudimentary query optimization at a smaller scale - say up to a few million rows. Can I also assume concurrency won’t be a consideration just yet? Yes those assumptions are fine

Your Answer

OK, before you read on, consider how you might answer this. Take a second and put down your phone/ipad/laptop (or just look away from the monitor) and practice answering this. It’s entirely likely you might not know anything about database indexes which is just fine… but then again you actually might know some basics about how they work. It’s all data structures and algorithms after all! If you’re totally in the dark on this that’s OK, read on just for fun to see how you might answer something more relevant to you.

The Simple Version

Most databases have some form of indexing that allows for faster searching and querying. Some databases, like PostgreSQL, allow you to choose what type of index you want to use - such as HASH, BTREE, GIN or GIST (among others). Each of these stores data in a particular way. Other databases have only a single type but they all do the same thing under the hood: pre-load some data into a data structure which is optimized for a search algorithm later on.

The simplest example which would answer the question above is to store the data pre-sorted so you could use something like binary search later on. For instance, you might want to index a users table on their email address. The index would sort all of the user records based on email in the background and you could then query that index using binary search.

The More Complete Version

An answer that shows that you know your data structures and algorithms (which I hope you do by now!) would be to use something like a balanced binary tree to store the data with depth-first search for the query. Do you have other ideas on this? Great! However… Samantha wants to know how much you know databases. It might be tempting to go on about index theory but resist the urge… no one wants to work with a showoff!


The goal with this scenario is not to prove how much you know about databases, but to see how an intimidating question such as “can you build an indexing engine” can be broken down to an acceptable answer if you:

  • Make sure to ask questions
  • Calm yourself if you feel out of your depth
  • Remember your data structures and algorithms - the answer might be right there!

And finally - if you really know an answer, show it by being as concise as you can.


The Code

Code for this video (and for the entire series) can be found in the production GitHub repo.

  • Let's Get Started!

    This is a quick introduction to what this production is all about and what (I hope) you'll learn. Please watch it. There are some supplies you'll need to get and some mental space you'll need to clear if you're to get the most out of this production.

  • Review: Big-O Notation

    It's a thing you'll need to know if you plan to get through any interview. It's actually quite useful and in this post I'll hopefully make the case that's it's simple as well.

  • Review: Data Structures

    Let’s review data structures! It’s fun and exciting! In this section you’ll review and then build your very own data structures along with me! I’ll be using JavaScript to code mine, but feel free to use whatever language you like.

  • Review: Algorithms and Strategies

    Now the fun starts! We have our data structures and now we need to focus on the algorithms we’re likely to be asked about in an interview. But that’s not enough! What you really need to do is to be able to tease out the strategy that you’ll need to use in combination with your data structures and algorithms. That will help you answer almost any interview question.

  • Your Interview Mindset

    You're not going to make it through this section unless you have the right frame of mind. Simply wanting a job isn't enough - you need to calm yourself and focus.

  • Getting Past the Screener

    Before you get to the white board you'll likely have to pass an introductory screening interview - a short call (or in person meeting) to suss out what you know.

  • Databases: How Would You Implement...?

    Screener questions tend to have a bit more broad and, typically, based on your experience... sometimes not. Think about how you would answer this one about databases.

  • JavaScript: What Time Is It?

    Sometimes a potential interviewer will 'cosplay' with you - pretending that you work at the company and seeing how you might solve one of their realworld problems.

  • Understanding the Mechanics of The Interview Question

    Don't hate me... and PLEASE don't skip this video. Let's walk through Fibonacci together as a way of understanding how interviewers will flex and extend a single question to dig into your brain.

  • Practice: Stock Price Calculation

    This is a real interview question that comes straight from Amazon. You'll be writing this one by hand so remember your strategies.

  • Practice Question: Product of All Numbers But This

    A straightforward problem that, at first, seems daunting - until you break it down and apply the strategies you know.

  • Practice Question: Smallest Range of K Lists

    This is a tough one and you might struggle with it some. If you do, practice speaking aloud to get help from the interviewer.

  • Jon Skeet: Creating a C# Linter

    Another traditional question: create a linter for X language. This is more about parsing a string and remembering where you are within it.

  • Jon Skeet: Falling From Great Heights

    Not all interview questions have to do with code - some are purely logical, designed to see how you would try and solve a really, really hard problem. Skeet couldn't figure this one out...

  • Jon Skeet: Binary Tree Troubles

    It's a rule of interviews: 'ask a binary tree question' and that's exactly what I do in this video with Jon. Traversing in different ways, trying to balance along the way.

  • Jon Skeet: Creating a Queue from Scratch

    This is another very popular question from larger companies, and requires you to be familiar with linked lists, arrays and stacks.

  • Jon Skeet: Anagram Checker

    Let's end things with one that's a bit simpler - creating a string walking function that evaluates anagrams.

Watch Again



Next Up