Millions of résumés never make it past the bots. One man is trying to find out why.

Mint

23-06-2025

U.S. job hunters submit millions of online applications every year. Often they get an automatic rejection or no response at all, never knowing if they got a fair shake from the algorithms that gatekeep today's job market.
One worker, Derek Mobley, is trying to discover why.
Mobley, an IT professional in North Carolina, applied for more than 100 jobs during a stretch of unemployment from 2017 to 2019 and for a few years after. He was met with rejection or silence each time. Sometimes the rejection emails arrived in the middle of the night or within an hour of submitting his application.
Mobley, now 50 years old, noticed that many of the companies he applied to used an online recruiting platform created by software firm Workday. The platforms, called applicant tracking systems, help employers track and screen job candidates.
In 2023 Mobley sued Workday, one of the largest purveyors of recruiting software, for discrimination, claiming its algorithm screened him out, based on his age, race and disabilities. Mobley, a Black graduate of Morehouse College who suffers from anxiety and depression, said the math didn't add up.
He says he applied only for jobs he believed he was qualified for. 'There's a standard bell curve in statistics. It didn't make sense that my failure rate was 100%," said Mobley, who has since gotten hired and twice promoted at Allstate.
His suit is now emerging as the most significant challenge yet to the software behind nearly every hiring decision these days. Last month—after several failed challenges by Workday—a federal judge in California said Mobley's age-discrimination claim could proceed, for now, as a collective action. The ruling opens the door to millions of potential claims from job seekers over the age of 40.
While the judge has ruled that Workday didn't intentionally discriminate against Mobley, she left open the door for him to prove that Workday's technology still had the effect of penalizing him because of his age. She hasn't addressed the race and disability claims.
Mobley still has a tough case to prove, and the suit may go through years of legal wrangling. Yet the case could force Workday to part the curtains on how its algorithm scores applications, a process that has remained a black box since job searches began moving online decades ago.
'Hiring intermediaries have pretty much been excused from regulation and they've escaped any legal scrutiny. I think this case will change that," said Ifeoma Ajunwa, a professor at Emory University School of Law and author of 'The Quantified Worker."
Workday says Mobley's claims have no merit. It said its software matches keywords on résumés with the job qualifications that its employer-customers load for each role, then scores applicants as a strong, good, fair or low match.
While employer clients can set up 'knockout questions" that lead to automatic rejections—for example, asking if a person has legal authorization to work in the U.S. or is available for weekend shifts—the software is designed so employers make the final decisions on candidates who make it through the initial screen, Workday argued in court filings.
'There's no evidence that the technology results in harm to protected groups," the company said.
Before his job search, Mobley's career path hadn't been smooth. He was laid off in the recession that followed the Sept. 11, 2001, terrorist attacks and again after the housing meltdown in 2008. After that, he left finance and transitioned to what he viewed as a more recession-proof career in technology, earning an associate degree in network system administration.
Still, steady jobs were hard to come by. He spent a year as a contractor at IT firm HPE, hoping the stint would turn into a permanent position. Mobley said he was let go, and he later joined a lawsuit against HPE alleging age and race discrimination. The case was settled in 2020. HPE declined to comment.
That job loss led to two years of unemployment, starting in 2017. He applied to more than 100 jobs and found himself on Workday's recruiting platform over and over.
He didn't get a single interview, let alone a job.
Soon, Mobley felt he discerned a pattern. 'It dawned on me that this must be some kind of server reviewing these applications and turning me down."
He worried that hiring software screened him out because it picked up on his age and race through details on his résumé or that it detected his anxiety and depression through personality tests he took as part of some job applications.
The frustrations of the job search weighed on his emotional health, credit and retirement savings, he said. He stayed afloat by driving for Uber and working short-term jobs.
Mobley eventually did find a job, the old-fashioned way. In 2019, he said, a recruiter for Allstate called him. A phone screen led to an interview with a hiring manager and then an offer. He is now a catastrophe controller, managing the workflow of customers' property and auto damage claims.
Mobley said he suspects Workday's software flagged his profile, essentially blackballing him across its entire system, regardless of which company he applied to.
Workday disputes that idea, and HR technology experts are skeptical of the theory. Employers customize recruiting software with their own criteria, they say, creating closed systems that shouldn't theoretically speak to each other.
But there is evidence that underlying scoring algorithms can shut out certain job seekers, said Kathleen Creel, a computer scientist at Northeastern University who has been following the Workday case. That might happen, she said, through mechanical errors such as misclassifying a previous job title, or by incorporating more complicated algorithmic mistakes that penalize members of a single group or people with certain combinations of characteristics.
Such scoring systems can disadvantage qualified workers, according to researchers at Harvard Business School, who have found that the systems effectively screen out millions of workers by scoring them low for all kinds of reasons, such as having gaps in their résumés or not matching every qualification listed on a lengthy job description. The researchers didn't test for illegal discrimination, such as discrimination based on age, gender or race.
Since 2022, Workday has built a team focused on ensuring its products meet ethical artificial-intelligence standards. 'Our customers want to know, can I trust these technologies? How were they developed?" Kelly Trindel, who leads the ethical AI team, said at a conference this month at New York University Law School.
Still, the company has fought some efforts to regulate automated hiring tools. In 2023, a New York City law went into effect requiring employers that use technology like chatbot interviewing tools and resume scanners to audit them annually for potential race and gender bias, and then publish the results on their websites. When the bill was proposed, Workday argued to loosen some of the rules.
If Mobley succeeds, software companies and their customers may be required to do more due diligence and disclosure to ensure they don't enshrine bias. Employment lawyers say any finding of liability could open the door to job seekers also suing employers who use them.
'This isn't a personal vendetta," Mobley said. 'I'm an honest law-abiding person trying to just get a job in an honest way."