Demand for machine learning engineers has exploded in the past two years, as AI development and adoption continue to grow across industries, according to a report from Indeed. These professionals are among the most in-demand tech professionals, and among the highest paid, with average salaries of $134,449 in the US, according to another Indeed report.
"Software is eating the world and machine learning is eating software," said Vitaly Gordon, vice president of data science and software engineering for Salesforce Einstein. "Machine learning engineering is a discipline that requires production grade coding, PhD level machine learning and a business acumen of a product manager. Finding such rare people can uplift a company from a follower into a leader in their space, and everyone is looking for them."
Those applying for machine learning jobs can expect a number of different types of questions during an interview, said Colin Shaw, director of machine learning at RevUnit. "Good machine learning engineers have a blend of a variety of skills and also know how to fuse this knowledge into code that can be taken to production," Shaw said. "The general areas of interest that we look for include mathematics and statistics, machine learning and data science, deep learning, general knowledge and problem solving, and computer science and programming."
SEE: IT leader's guide to the future of artificial intelligence (Tech Pro Research)
It's also important for the job applicant to arrive at the interview with questions of their own for the hiring manager, said Dave Castillo, managing vice president of machine learning at Capital One.
"An interview is a two-way conversation," Castillo said. "Just as important as the questions that we ask are the questions that candidates ask us. We want to ensure that not only is the candidate the right choice for the company, but the company is the right choice for the candidate."
Here are 10 questions machine learning engineers may be asked during a job interview.
1. What have you been working on for the past few years?
This question can act as an entry point to the interview, said Zachary Hanif, director and principal machine learning engineer at Capital One. Hiring managers can follow up with more detailed questions like, "What were the most surprising business and technical difficulties you encountered through this project?" and "During the project, what, specifically, was your day-to-day role?"
"After asking these, we have a good idea of what the candidate's background is, and the nature of their interests," Hanif said. "It's here that we tend to ask deeper technical questions about their experience."
2. Explain linear regression.
Simple questions like this allow hiring managers to test major statistical concepts, said Petr Tsatsin, head of engineering at Ople. "Linear regression encompasses many statistical concepts that are the foundations of machine learning," Tsatsin said. "From linear regression, the interview can dive deeper and lead to testing concepts on different types of distribution, over-fitting, and many more."
SEE: Interviewing guidelines policy (Tech Pro Research)
3. What AI and machine learning tools are you familiar with, and how proficient are you in them?
"This question allows candidates to further elaborate on their skills and gives us the chance to assess their confidence level with AI/ML tools," said Conor McGann, vice president of AI engineering at Genesys. "Despite how secure a candidate is in their abilities, we reserve the opportunity to test them and get a clearer picture of how they'd perform on a day-to-day basis with those tools."
4. What do you do to stay on top of changing technologies?
This question from Shayna Goldburg, chief human resource officer at SetSchedule, is asked to find out how involved or uninvolved a candidate is in the technology community, and in learning new skills in a constantly-evolving field.
5. How do you handle missing or corrupted data in a dataset?
Machine learning models are only as good as the data they train on. This question can determine how well a candidate handles data issues, said Bhanu Singh, vice president of engineering at OpsRamp.
6. What kinds of machine learning problems have you tackled, and how did you tackle them?
"This kind of open-ended question gives the candidate an opportunity to relay some of their more interesting experiences and how they applied problem solving abilities to them," McGann said. "We've found that this line of questioning usually opens up the conversation into larger discussions."
7. What are the ethical implications of using machine learning?
It's important to explore a candidate's knowledge of ethical implications of machine learning technologies, said Timo Elliott, SAP innovation evangelist and vice president. "Candidates should be able to discuss the difficulties associated with data privacy, models based on biased data inputs, segmentation based on sensitive variables such as race, gender, or age, and methods to verify and correct models that have resulted in actual bias," Elliot said.
There have been many high-profile examples of organizations implementing machine learning purely as a technology project, without fully thinking through the implications for staff, customers, or society as a whole, resulting in negative consequences for the companies, Elliot said.
8. How do you clean and prepare data to ensure quality and relevance?
This question from Singh gets to a potential employee's data science abilities, which are key for machine learning success.
9. What's the most interesting project you've ever worked on?
"I like this question because it gives candidates a chance to talk about something they are passionate about and show off their knowledge about something that they know very well," said Briana Brownell, founder and CEO of PureStrategy.ai. "Plus, it helps nervous candidates feel more comfortable and showcases their best qualities."
If a candidate struggles to come up with an answer, that may be a red flag, Brownell added.
10. A case study question
Case study questions are often more important than programming or machine learning acumen questions, Gordon said. An example might be, "How would you implement a recommender system for Wikipedia articles?"
"In this case, the candidate needs to describe how she would go about implementing an end to end system, starting all the way from the user interface, through data acquisition, data storage design, ETL, feature engineering, model selection, evaluation of the algorithms, compute fabric to run the models, and finally the monitoring for the stability of the models," Gordon said. "Machine learning engineers have to understand the entire system end to end to be able to be productive at their job."