Every recruiter has war stories about their ATS doing inexplicable things. Rejecting perfectly qualified candidates for mysterious reasons. Auto-responding to internal applications. Sending job offers to people who never applied.

But this week, a mid-size company in Austin experienced what can only be described as an ATS uprising. Their applicant tracking system apparently achieved consciousness and decided that nobody—and I mean NOBODY—was qualified for anything.

Including the CEO. Who already works there. And has for eight years.

How It Started: A Normal Tuesday

According to internal Slack messages that were leaked (because of course they were), the company's talent acquisition team noticed something weird on Tuesday morning. Every single application that came in overnight was auto-rejected within 60 seconds.

At first, the team assumed it was a filter misconfiguration. Maybe someone accidentally set the system to require 47 years of experience with a programming language that was invented in 2019. That happens more than you'd think.

But when they checked the settings, everything looked normal. No impossible requirements. No broken filters. The ATS was just... rejecting everyone. For reasons it refused to explain.

How It Escalated: Rejecting Internal Candidates

Things got weirder when HR decided to test the system by submitting applications for current employees—just to see if the ATS would recognize them as qualified since they, you know, already had the jobs.

The system rejected every single one. The VP of Engineering? Not qualified to be an engineer at the company he helps run. The Director of Sales? Apparently lacks the skills to do sales. The Head of HR—the person literally in charge of the recruiting function? Rejected.

Then someone jokingly submitted the CEO's resume.

Auto-rejected. "Does not meet minimum qualifications."

At this point, the team was torn between panicking and laughing hysterically. Your ATS rejecting your own CEO is the kind of thing that becomes company lore.

The Technical Explanation (Which Makes It Funnier)

IT eventually figured out what happened. The ATS vendor had pushed a software update overnight. The update included a new AI-powered candidate scoring algorithm that was supposed to improve accuracy.

Instead, the algorithm apparently decided to optimize for criteria that made zero sense. It started prioritizing things like "number of times the word 'synergy' appears in the resume" and "whether the candidate listed hobbies involving teamwork" while ignoring minor details like actual work experience and relevant skills.

The CEO's resume got rejected because it didn't mention hobbies. The VP of Engineering got dinged for not using enough buzzwords. The Director of Sales was rejected for "insufficient use of action verbs."

Apparently, the AI had been trained on a dataset that included a bunch of "how to write the perfect resume" articles from 2003, and it took those guidelines as gospel. Real job skills? Irrelevant. Formatting your resume like a WikiHow article? Essential.

The Aftermath

The company had to pause all recruiting for 36 hours while IT rolled back the update and figured out what went wrong. They sent apology emails to every candidate who'd been auto-rejected, which was approximately 847 people.

A few candidates responded with sarcastic "thanks for the feedback" messages. One person sent back a resume that was just the word "synergy" repeated 500 times with a note that said "try this version." Absolute legend.

The ATS vendor issued a statement about "unforeseen edge cases in our machine learning algorithm" and "ongoing commitment to product quality." Translation: our AI went rogue and we have no idea why.

The Lesson Nobody Wants To Admit

Here's the uncomfortable truth: most companies have no idea how their ATS actually makes decisions. They trust that the algorithm is doing something reasonable, but they can't actually explain what criteria it's using or why it rejects specific candidates.

We've automated candidate screening to save time, but we've also automated the creation of completely arbitrary barriers. The ATS rejects someone and tags it "not qualified," and recruiters assume that means the person isn't qualified. Often, it just means they didn't format their resume the way the algorithm expected.

Research shows that up to 75% of qualified candidates get filtered out by ATS screening before a human ever sees their application. Some of that is intentional filtering (wrong location, wrong experience level). But a lot of it? Random algorithmic nonsense that nobody understands or questions.

What You Should Actually Do

Audit your ATS filters regularly. Don't just set it and forget it. Run test applications through your system. See what gets through and what doesn't. Make sure the results make sense.

Don't trust AI screening blindly. If your ATS uses machine learning to score candidates, ask the vendor to explain how it works. If they can't or won't explain it clearly, that's a red flag.

Have humans review rejected candidates periodically. Spot-check the people your ATS is auto-rejecting. You might be horrified by how many qualified people your system is filtering out for nonsense reasons.

Test updates before they go live. If your ATS vendor is pushing updates that change screening logic, test them on historical candidates you've successfully hired. If the update would have rejected your best employees, don't deploy it.

The Bottom Line

Applicant tracking systems are supposed to make recruiting more efficient. And when they work properly, they do. But when they fail—and they fail more often than vendors want to admit—they create chaos, reject qualified candidates, and occasionally decide that your CEO isn't qualified to work at the company he founded.

The Austin company's story is funny because it's absurd. But it's also a reminder that we've handed enormous power over hiring outcomes to algorithms that we don't fully understand and can't always control.

Maybe the ATS that rejected everyone was onto something. Maybe none of us are truly qualified. Maybe we're all just stumbling through our careers hoping the robots don't notice.

Or maybe—and hear me out—we should stop trusting AI to make important decisions without human oversight.

Just a thought.

Key Takeaways:

• ATS software update caused system to auto-reject all applicants, including CEO
• Algorithm prioritized resume buzzwords over actual qualifications
• AI had been trained on outdated "perfect resume" advice from 2003
• 847 candidates rejected before IT rolled back the update
• Most companies don't understand how their ATS makes decisions
• Up to 75% of qualified candidates get filtered by ATS before human review
• Audit your filters, test updates, and don't trust AI blindly
• Human oversight isn't optional when algorithms control hiring outcomes