Just about every industry is being reshaped by AI. That means there’s a good chance the company you work at is or soon will be building AI as part of its products, services, or internal tools. With so many companies around the world looking for the talent to compete in this space, how do you find and hire people who can build AI systems?
Naturally, you’ll find AI talent in emerging fields such as self-driving vehicles and smart speakers. But even with a good sense of where AI is being implemented, you may still be losing sleep over how to find and hire AI talent, or how and when to retrain your existing teams. Here are three areas to tackle first so you don’t get left behind.
Understand the skills you need
In LinkedIn’s annual report of the fastest-growing jobs of 2018, we found that machine learning engineer and specialist were among the top five, showcasing that the search for this talent is at an all-time high.
When hiring this type of talent, one approach is to identify and hire candidates with the underlying hard skills needed to develop skills in machine learning. Our data show that skills for machine learning talent include deep learning and Python, which are similar skills for a data science specialist. These data scientist professionals are already on a path to add computer science, data visualization, and then AI skills. Therefore, consider expanding your candidate talent pool to include keywords such as “statistics” and “data analysis” to your search for upskilling opportunities.
Learn the best way to look for soft skills
Most candidates aren’t going to come to you with the full spectrum of skills required to implement AI so you’ll also want to look for those hard to find soft skills. In fact, 57% of leaders say soft skills are more important than hard skills. These skills, like leadership, communication, collaboration, and time management, will help people adapt and pick up important technical skills quickly. Of course, this means that willingness to learn is key.
To screen for soft skills, consider asking specific problem-solving questions during the interview process. Many companies find that providing a challenge, then quickly changing it to see how a candidate adapts, showcases how they approach situations.
Cast a wider net
Our data show that software engineers don’t apply for jobs as often as other professionals, but they’re 12% more likely than other professionals to respond to a recruiter about a new job opportunity if you reach out to them.
Even so, if you’re located in a tech hub like Silicon Valley, you’ll have less competition if you’re willing to look elsewhere for talent. Many engineers live in so-called hidden gem markets like Los Angeles, Dallas/Fort Worth, and Philadelphia, which have a large supply and relatively low demand for these workers. Data-driven recruiting tools, such as LinkedIn Talent Insights, help to survey talent pools in specific areas and match skills with the number of jobs to identify these hidden gems.
Beyond geographic location, you should also consider looking outside your industry. Our research shows that 72% of professionals with machine learning skills who changed jobs last year also changed industries. Thirty-three percent of these professionals are in higher education and research, 26 percent are in the software and internet industries, and 5 percent are in finance and banking. Our data shows that higher education and research is the best source of machine learning talent most likely to pivot into another industry. More than one in three such people on LinkedIn started their careers in higher education and research, primarily serving as researchers.?
Once you have a pool of strong candidates, be open to new ways of testing for AI or related skills or the potential for an AI role. More than 50% of respondents in LinkedIn’s 2018 Global Recruiting Trends survey found that job auditions–which involve putting candidates on actual projects–are useful for hiring.
As an example, Citadel teams up about 100 students in a competition with a cash prize. The students use data to solve business problems, while recruiters assess how participants think and behave. The program resulted in dozens of hires.
Retrain your workforce
Our data show that the supply of AI candidates simply can’t keep up with demand. Data scientist roles, for example, have grown more than 650% since 2012. Hundreds of companies are hiring for these roles, but only 35,000 people in the U.S. have data science skills.
To close the gap, begin by identifying where demand is accelerating the fastest in your company and where you see potential areas that need a boost. Then create a strategy to reskill certain employees. Given that the average shelf life of skills is less than five years, it helps to shift your mindset from one-size-fits-all training to personalized, continuous learning. To get ahead of these challenges ourselves, we’ve launched an AI academy for our engineers that covers the basics of implementing AI.
The demand for AI skills is here to stay. Companies and managers that focus on soft skills that drive innovation, think outside the box to recruit AI talent, and develop a focus on continuous learning will be best equipped to meet any new, unforeseen challenges in an ever-changing workforce.
John Jersin is vice president of Product Management, Talent Solutions, and Careers at LinkedIn.