Machine Learning System Design Interview

Product Information

Deep Reinforcement Learning for Online Advertising in Recommender Systems (TikTok) - this uses Reinforcement Learning for recommendations Alexey: Okay. So let's go to the questions. We have quite a few of them. The first question we have is, “What are the typical components of a machine learning system? And what percentage of it are machine learning algorithms?” ( 47:52) This book is the result of the collective wisdom of many people who have sat on both sides of the table and who have spent a lot of time thinking about the hiring process. It was written with candidates in mind, but hiring managers who saw the early drafts told me that they found it helpful to learn how other companies are hiring, and to rethink their own process. Valerii: Okay, let's try to determine the disparity between those two. First of all, when you're asked to do a system design interview, you're usually asked about data structures, about different server-side components, like “What are the databases? What is the amount of data that will be processed? What is the write throughput? What is a read throughput? How would you work with a cache? How would you work with load balancing, sharding, splitting?” etc, etc. So it's basically software engineering. ( 13:58)

This repo is written based on REAL interview questions from big companies and the study materials are based on legit experts i.e Andrew Ng, Yoshua Bengio etc. Alexey: Okay. I don't think we have a lot of time for more questions. There is an interesting question from Vijay, which is about, “What is the best way to validate the model performance in production? Do we need humans for that or are there other ways?” ( 57:23)Valerii: Did you use the data scientist profile, because I told you that I don't like “data scientist” in my job title? I find it awful and terrible. So you’re just nudging me in my pain point. ( 58:47)

You can also make use of other creative data collection techniques. For example, you can build a personalized experience in your product by collecting data from users. If you’re working with a system that uses visual data, such as object detectors or image segmenters, you can use GANs (generative adversarial networks) to enhance the training data. Other things to consider include: Valerii: Yeah, it's a good taxonomy. It's a good taxonomy. It's a good book. If it didn't reveal anything new to me, it doesn't mean it's a bad book. It just means that it's my problem. ( 53:24)Alexey: Yeah, exactly. Okay. Maybe one last question. It seems like you have a very solid data science profile, from Grandmaster at Kaggle. That's pretty solid. ( 58:35) Valerii: British HR would never write you that. They would say something like “Alex, it was wonderful. It was brilliant. There was just that slight miscommunication.” Something like that. They'll never tell you that you completely failed. Never. ( 28:10)

Often, the interview question may be about the real product the company develops (as opposed to some theoretical situation.) That means you’re expected to make some common sense assumptions. Does Facebook need to support multiple languages? There’s no need to explicitly ask that, but you should call out any assumptions you’re making. You can also clarify any preconceived notions you have about the solution. For example, you might call out that it seems important to be able to show recent posts from friends, but you could get clarity on what ‘recent’ means, posts from the last 5 minutes or the last 24 hours. What’s the oldest content you would recommend? High Level Design Alexey: That's quite a lot of information. I was trying to process this. That's quite a lot of things. So this was an example of machine learning system design. The interview starts and then the person – the interviewer – asks you, “Let's design a system for detecting fraud.” And then you probably ask this person a few questions and then you do this information dump on that person, right? ( 20:33) One of the best resources that focuses on the first principles behind designing ML systems for production. A must-read to navigate the ephemeral landscape of tooling and platform options." - Goku Mohandas, Founder of Made With MLLooking to start a career in data science and AI and do not know how. I offer data science mentoring sessions and long-term career mentoring: Alexey: Okay. So we do this, and then you also mentioned A/B tests. We define a metric, and then we say how exactly we are going to measure this metric. What do we do next? ( 44:01)

Once the model is launched, what other ops work will there be? How can we monitor the model to make sure it’s healthy and what operations will we have to do to keep the model performing well. What happens if we want to update our features? Leveling Alexey: Yeah, I guess the answer might be just being a practitioner? Because models don't live in isolation, right? ( 59:37) Alexey: Perhaps if you cover all these parts during your system design interview, you're already in quite a good position. Right? ( 46:02)I’m a SWE, ML with 10 years of experience ( Linkedin profile). I had offers from Google, LinkedIn, Coupang, Snap and StichFix. Read my blog. After asking questions, you should carefully choose your system’s performance metrics for both online and offline testing. These metrics will differ depending on the problem your system is trying to solve. One of the most important design decisions is whether the system is real time, pre calculated batch or some hybrid. Real time systems limit the complexity of the methods available while batch calculations have issues dealing with staleness and new users. Machine learning systems design is the process of defining the software architecture, infrastructure,