
According to a recent survey, more than 30% of U.S. companies reported the use of AI and machine learning in their decision making and almost 55% of the companies expect to roll out major initiatives within the next few years. This has created a very high demand for AI & ML professionals who are qualified enough to deal with the complex data projects evolving in almost all major organizations. As the demand increases, the supply of talent in this space is in short supply. Many universities across the globe has launched specialized courses and are already producing a lot of wannabe AI and ML professionals, but the fact remains that the real talent is scarce and will remain so for the foreseeable future. Companies are also reporting a mismatch between skills and expected compensation, which is obviously fueled by the huge demand supply gap that exist today. It is thus imperative that companies put a special emphasis on attracting on retaining talent especially when they are competing with Google’s and Amazon’s pf the world for the same talent.
In this article, we share with you our best practices for recruiting AI & ML candidates, including how to define the specific roles, various talent acquisition strategies to attract the best talent, assessment and selection techniques , and finally how do you retain the talent once they are on-boarded.
One of the most common mistakes, hiring managers make is when they try to hire a Unicorn. A Unicorn is someone who has top rated skills in data engineering, programming, traditional machine learning and modern Ai techniques such as deep learning. This Unicorn also comes with a strong business acumen and is excellent at people skills, communication skills and political skills. You get the picture; such Unicorn does not exist. One must be able to look at the specific job specification and define the job description which is very specific to most critical requirements of the job. Trade-offs must be made at the time of writing job description. It makes sure that you do not turn away good talent because you are asking for too much. If the job demands good experience in traditional machine learning techniques, don’t ask for must have deep learning skills or data engineering skills or model deployment skills. Hire for what you need now and then focus on developing that talent for future needs once the candidate is hired. So, think about the role carefully. Is it a ML engineer, a programmer, a MLOps engineer, a data engineer, a business analyst, a data analyst, a DL programmer or research scientist? In other words, define the role carefully to attract the right talent for the job including soft skill requirements and domain knowledge requirements.
Be careful with putting too much emphasis on domain expertise as one of the requirements. Yes, someone with experience within your industry is ideal, but is it a good decision? Hiring an average candidate with industry experience is far inferior to hiring a top talent from other industry and training that person to bring him up to speed with your business processes. Hiring from other industry has also been a good way of bringing fresh perspective as these candidates walk in with no preconceived notions and beliefs about your business.
World of ML and Ai is changing fast. New techniques, tools and algorithms are being released on almost daily basis, so its very important for the candidates to show strong learning skills. it is also very important to keep up with the latest research breakthroughs, so an aptitude to constantly scan the research papers and try out new things is very important.
Innovation and creativity are another important quality to look for when hiring for AI & ML roles. Someone with just bookish knowledge is not going to be able to solve the complex problems for your business. Your candidate should be able to come up with out of the box thinking for every problem they are trying to solve. Creativity is a very hard skill to acquire, a strong aptitude is required, so when hiring for these roles, you must check for how they can think on their feet. One way to do that is to give the candidates impossible situations. You are only looking for how they respond to such situations and how innovative they can get. A candidate with very limited creative capapbility will draw blanks and that’s a red flag to look for. A re4lated skill to look for is “curiosity”. A curious candidate will not be afraid to ask the right questions and develop deep understanding of the business processes.
In our experience business acumen is a very important skill to have for AI and ML candidates. They need to have a passion for understanding the business problem holistically. One of the common mistakes hiring managers make when they only check for technical skills. Remember, you are looking for someone who can solve business problems and not just math problems.
Finally, AI & ML candidates must have a strong data analysis aptitude. Almost 80% of any AI/ML problem in the corporate world is about gathering, transforming and analyzing datasets. You cannot overlook this critical skill. Research has shown that candidates with ultra-strong data analysis skills tend to perform the best in the long run and add the most value. We strongly recommend testing their data analysis capabilities as part of the hiring process.
Once you’ve recruited a strong team of AI & ML experts, another problem arises: how to keep the talent you’ve just spent months recruiting. If you are not careful, you will end up losing your top talent within a matter of few months. Demand for Ai and ML professionals is so strong that they will be getting calls from external recruiters on a weekly basis with very attractive offers. You must ensure that their compensation is in line with the market, there is nothing you can do if the market rates are 40% higher than what you are paying. Regular communication and early setting of expectations are crucial steps to ensure that they understand their roles and are committed to achieving the goals laid out for them. Lack of communication and unrealistic expectations is a perfect recipe for creating frustrated employees leading to attrition. You must develop a culture of constant learning, training, and risk taking.
AI and ML professionals are hard to attract and retain in today’s highly competitive market. Its important to keep the above points in mind if you want to build and maintain a healthy AI and ML team.