5 Job Interview Tips From An Airbnb Machine Learning Engineer
In this article
Springboard mentor Chirag Mahapatra started his career as a software engineer at Goldman Sachs before taking the leap into data science and machine learning engineering, where he worked at Amazon, Trooly, and eventually Airbnb.
In this blog, Mahapatra shares insights into what hiring managers often look for in machine learning engineers and tips on how to prepare for an ML engineering job interview.
Tip 1: Know your algorithms
If you’ve gotten as far as the interview stage for a machine learning engineering role, you’ve probably had to show recruiters and hiring managers that you have a solid understanding of AI and ML algorithms and can put them into practice. Having a clear notion and solid grasp of the required algorithm is recommended whether you work in the machine learning department or the data science industry. But you should also be prepared to talk about it, according to Mahapatra, who said from personal experience that interviewers want to know that a candidate can explain what machine learning is and discuss the algorithms and tools deployed in different projects.
“There will be an interview about your experience where you are asked, ‘Have you actually built machine learning systems in the past?’ and ‘What could [we] have [done] better?” Mahapatra said.
At Airbnb in particular, some of the questions will be programming and algorithm-based, so even if a candidate has deep knowledge and experience in using ML algorithms, they should practice articulating their ideas and include examples of projects where they can explain their algorithmic choices.
“I think what Airbnb likes to see is a lot of execution capability,” Mahapatra said. “That is, you’re actually able to write code quickly and then be able to implement it and make sure it’s working in the [time allotted for] the interview.”
Tip 2: Be curious and willing to learn
It’s one thing to arrive at a company with a strong technical skillset and machine learning engineering experience. It’s another to have all of the above and a curious, open mind that’s open to learning, experimentation, and improvement. Recruiters often use the interview stage of the screening process to identify candidates who have the right attitude, ambition, and creativity to join a company, so it’s worth highlighting your own curiosity and interest in learning when you are being interviewed.
“We are looking for people who are curious and who are willing to learn,” Mahapatra said of Airbnb. “Some background in machine learning, like building machine learning systems, is really well appreciated. [We also look for] skills to actually ensure that you can make the system work in production and a good understanding of different experimentation frameworks.”
Tip 3: Highlight your specialization
No recruiter or hiring manager expects a machine learning engineer or a data scientist to be an expert in everything, according to Mahapatra. Instead of stretching yourself thin trying to prove that you can do everything, he suggests highlighting the things you know really well.
“Machine learning is a field which is growing really fast. There’s no way a person can be good at everything. It’s just impossible,” he said. “We really like candidates who are really specialized—they know one thing really well and can do that [one thing] well.”
That one thing might be building training infrastructure, or natural language processing, or even the building of certain models that can go into production—focusing on just one thing and specializing in it can be an asset.
Tip 4: Be ready to explain everything—and have patience
As machine learning and artificial intelligence play a bigger role in ordinary people’s lives, it’s important for the engineers behind the algorithms to be able to explain their models, decisions, and the flow-on effects of their algorithms. This is something that hiring managers will be looking for, according to Mahapatra, who added that the ability to explain a model is an increasingly valuable skill.
Many machine learning engineers are able to build models that help organizations meet their metrics and goals, Mahapatra said, but those same engineers often struggle to explain how their models arrived at certain decisions—this simply doesn’t cut it anymore.
“So, I think there’s going to be a lot more work, especially in industry, around explainability,” he said. “Especially like right now, machine learning is getting used in more critical workflows. Earlier it was just things like recommendations and ads… but today, when you’re making decisions on people’s lives, then it’s much more important that you are able to explain why that decision was made.”
Tip 5: Experiment with tools and modeling
Similar to the first tip about being able to both show and discuss your experience with machine learning algorithms, hiring managers also want to hear about any experience with modeling, experimentation, and ML tools because they make up the bulk of a machine learning engineer’s job.
“The important thing to learn is, one, the modeling part, because you will be expected to do quite a bit of modeling,” Mahapatra said. “But apart from that, learning a lot about experimentation [and] how your model will actually be evaluated and practiced…that is something useful to learn.”
And when it comes to machine learning tools, Mahapatra recommends using several tools in projects that you can then discuss with hiring managers so that you can show that you not only know how to use them, but you can provide context and examples of where they can have the greatest impact.
“If you’re interested in natural language processing, then do like two to three national language processing projects,” he said. “That’s probably the path to take.”
Since you’re here…Are you a future data scientist? Investigate with our free step-by-step guide to getting started in the industry. When you’re ready to build a CV that will make hiring managers melt, join our 4-week Data Science Prep Course or our Data Science Bootcamp—you’ll get a job in data science or we’ll refund your tuition.
This post was written by Tracey Lien.