How Machine Learning Can Enhance Recruitment During COVID-19 and Beyond

Print Friendly, PDF & Email

There’s been no shortage of discussion in recent years about how the jobs of the future will involve professionals working alongside artificial intelligence. But AI isn’t just set to change the nature of jobs; it’s also slated to play an increasingly important role in helping recruiters determine whom to hire in the first place.

As with so many other innovations reshaping the economy and society, this trend has been accelerated by the coronavirus pandemic. Many organizations have conducted their hiring and onboarding processes virtually over the past year, with 67% of recruiters in a recent Jobvite survey indicating that they’re using video to interview job candidates and 40% saying they expect this to be the default solution in the future. As recruiters grow more reliant on technology for hiring, machine learning can augment the process by helping teams better assess candidates’ soft skills, personalities, and cultural fit.

Here’s why technology-enhanced hiring holds immense promise – and how to ensure that employers avoid some of the biggest pitfalls presented by these advanced tools.

Opportunity – and Risk

Flooded with applications and strained for time, recruiters have found hiring technology a vital asset over the past year, as it has helped them find candidates to fill positions with much greater efficiency.

Machine learning-based tools can scan resumes, assess candidates’ stated qualifications and skills, and even rank candidates according to how well they align with company values, culture, and desired soft skills – all critical factors in evaluating whether an applicant is likely to prove a successful hire.

Meanwhile, video-based interviews make it easier for recruiters to assess candidates on a level playing field, with the ability to review and compare answers both on recorded video and in transcripts. Being able to see and hear people help recruiters get a broader and better understanding of who the applicant is – specifically their soft skills, which may otherwise be overlooked or difficult to discern. From the candidate’s perspective, a video interview allows them to bring their true personality to light, allowing them to shine, while recruiters get to know future employees on another level, even before meeting them in person.

But just as introducing more technology into the hiring process creates valuable opportunities, it also opens risks. In an industry as sensitive and as critical to people’s lives and livelihoods as hiring, it’s important that hiring technologies – especially those based on AI – be implemented in a way that’s consistent with fundamental values like diversity and fairness.

It’s not just a legal consideration; it’s a moral and ethical one as well. If AI systems are inherently biased against historically disadvantaged groups – as some such systems unfortunately seem to be – it won’t matter if they make hiring simpler and more efficient, because they will ultimately make the workplace less fair, less conducive to equal opportunity, and probably less productive.

Systems that assess irrelevant or purely subjective factors have proven flawed, and if systemic biases are built into them, those biases will be replicated in the hiring process. In 2018, Amazon jettisoned its AI recruiting tool after the company’s machine learning experts discovered that it routinely penalized female candidates; applications featuring words like “women’s,” for instance, were consistently downgraded. More recently, a large recruitment technology provider removed a feature utilizing facial monitoring analysis due to problematic results and public backlash.

How can recruiters steer clear of similar scenarios?

Building Diverse Data Sets and Algorithms

To recruit a diverse workforce, it’s essential for hiring platforms to be diverse-by-design. That entails building diversity into every element of the platform – from data inputs to tagging to the teams that build the system itself, all the way through post-hoc monitoring to ensure that the system isn’t performing in a biased manner.

What does this look like in practice? Take video interview data: It must be sourced from different geographies, covering a variety of racial and ethnic backgrounds, personalities, speech types, and so on. If the team that tags the data is diverse, it will be more attuned to issues of diversity and fairness and will help combat the influence of unconscious biases that would otherwise make their way into the system. It’s important to not only guard against negative biases – against particular population groups, for example – but also “positive” ones, like those in favor of graduates from a limited set of elite institutions. 

In addition, a diverse and multidisciplinary team should take on the construction of the algorithms themselves. Unlike other AI models that may only require the skills of ML experts, behavioral psychologists must also play a key role in these systems’ inception and development in order to ensure the right factors are being taken into account. Full algorithmic transparency is crucial to combating bias and delivering a platform that will promote a dynamic, diverse, multifaceted workplace for all its users.

The complexity and sensitivity of hiring and the crucial impact it has on peoples’ lives require that companies take a thoughtful look at the solutions and systems they are considering – not only in terms of the impact they will have on the speed and efficiency of the hiring process, but also on the ability to create healthy company cultures.

As technology continues to play a greater role in hiring, now is the time for developers and HR professionals alike to commit to the next generation of recruitment – one that is more efficient, more diverse, more fair, and more vibrant.

About the Author

Benjy Gillman is an entrepreneur and technology expert with experience in building strong, cohesive teams. As myInterview’s co-founder and CEO, Benjy is instrumental in setting the strategic direction for the company and managing its success. Benjy holds a BBA from Macquarie University and a major in Property Development from the International College of Management in Sydney.

Sign up for the free insideBIGDATA newsletter.

Join us on Twitter: @InsideBigData1 –

Speak Your Mind



  1. Thanks for sharing this interesting info.

  2. The article you have shared is very informative and helpful, keep sharing!

  3. Thanks for sharing informative article

  4. This was a great piece on the power of machine learning to be adopted in recruiting. Hope to see much more informative blogs.

  5. Very informative, thanks for sharing it, looking forward to seeing more interesting blog post.

  6. very useful information! Thank you for sharing.

  7. Sounds Good!, Keep to share more article like this one!

  8. Such a valuable post! Thanks for sharing

  9. Machine Learning is having great future, well said

  10. Machine Learning is a very great future. Thanks for Sharing your information and it was really very helpful.

  11. Thank you for sharing your knowledge on such a rare topic.
    Machine learning is a great tool for recruiting professionals as it can help them assess the soft skills, personalities and cultural fit of candidates.
    Keep writing.

  12. As I’ve read the heading I was very curious to know about how ML can ease and enhance the recruitment process. By reading the article I can summarize that Machine-learning-based tools scan resumes and assess candidates skills. They can even rank candidates based on how closely they match company values, culture, or desired soft skills.