Dear Tech Recruiters (and by extension, the clients you work for),
A recruiter messaged me directly last week, which should have been a good thing. An actual person reaching out - wow!
Then they sent through the job ad itself: an AI-generated checklist so long and so specific that no real human could tick every box unless they were already doing the exact same job on the exact same stack at a direct competitor. The contrast between the personal touch and the impersonal wishlist was jarring, but sadly common in 2026.
The problem today is that AI has changed recruitment, and made it much, much worse. I want you to understand how this all looks to people like me.
First up - I've read enough job ads over the years to build up a "Hall of Fame" of recruitment nonsense, which I present to you now. My simple advice for all of these - do not do this.
Job Ads Hall of Fame
1. The vendor-specific must-have
Candidates must have direct hands-on experience in [Specific Platform].
A while ago I worked at a place which used a CMS I wasn't familiar with. But I've got 2 decades of experience in other CMS's, and 3 decades in tech. I asked a developer to talk me through the architecture, and as soon as he explained it, I could relate my existing knowledge to the new platform.
If you're in a leadership role, the skill is understanding the functionality, being able to define the problem, and knowing how to achieve the required outcomes. Functionally, every CMS, every CRM, every email system, every operating system, and every AI platform, does the same thing as its competitors. Learning one specific piece of software doesn't mean you understand the functionality and processes and decision making required to achieve the best outcome. Screening someone out for a technology leadership role based on a specific vendor is shooting yourselves in the foot.
2. The narrow sector obsession
Essential: minimum X years experience within [Specific Sector].
Modern software doesn't behave differently depending on which industry runs it. Sector context matters for some roles, but for most tech positions this requirement just shrinks the pool for no real gain. A technologist who's worked across sectors usually brings new ideas, concepts and processes that the narrow-industry-focused haven't considered. Screening someone out for a technology leadership role based on the sectors they've worked in previously is a nonsense.
3. The impossible experience requirement
10 years of experience as a pre-sales or engineering manager in relation to AI.
I've already covered this one: Google wanting someone with 10 years of AI experience. When I posted that on LinkedIn I got massive engagement, mostly the laughing emoji. It boggles the mind that this needs to be said out loud but - if your role requires more years of experience than the technology has existed, you are effectively telling us all exactly why we don't want to work for your company. Please apply some logic here.
4. The combination
Must have: [Specific Tool] certification plus X+ years in [Specific Sector].
This one combines several or all of the above. The only person who ticks every box is already doing this exact job, on this exact stack, in this exact sector, at a competitor. Which begs the question - why would they leave? And more importantly - why bother running an ad, when you know exactly who you're targeting - just pick up the phone and try to headhunt them!
5. The mystery salary
Salary: Competitive.
This tells me nothing, except that you're OK with me wasting an hour crafting an application for a role before finding out the role pays $50k less than my last role. A simple dollar range would save everyone some trouble.
6. The obviously AI-written ad
We are seeking a dynamic, results-driven professional who will leverage cutting-edge technology to drive transformative outcomes in a fast-paced environment — someone who can seamlessly wear multiple hats while championing innovation at every touchpoint.
You can spot these in the first sentence, and that em dash is a dead giveaway. It seems everyone wants a dynamic, results-drive professional and keyword keyword keyword these days. Please understand that when you post a job ad that sounds like a computer has generated it, you are telling us an awful lot about your organisation that maybe you're not intending to tell us.
Next up, we have all the ways candidates cop the raw end of the deal.
Bad Behaviour
- The keyword-blast. Recruiters who've clearly run your CV through their keyword-matching software, without reading it, and then send "opportunities" that have little to no relevance to your actual skillset or stated preferences.
- The silent rejection. A form email with no phone call and no feedback, when at senior level a 5-minute call costs almost nothing and leaves a completely different impression.
- No rejection at all. You apply and simply never hear back, not even the generic form email. We know you have the automated systems to send that email, it's one click, and that's bad but it's still better than nothing.
- The fake recruiter. A dodgy recruiter has basically stolen a job ad from a job board and started contacting potential candidates directly. Why do they do this? Unclear, but it gives the whole industry a bit of a bad odour.
- The junior who manages the boss's LinkedIn account. As a candidate you may assume you're connecting with the senior recruitment expert on LinkedIn who understands both sides of the industry and is best placed to help you find the perfect role. And you would likely be wrong. You're just as likely to be talking to the office junior who operates that account on their behalf, and the senior expert never sees anything you've said.
- AI interviews. Yes, the absolute best way to assess potential candidates is to put them in front of an AI bot as a first-round interview. Nothing speaks "cultural fit" and "positive work environment" like assessment-via-algorithm and the subliminal message that we're too busy to actually talk to you.
- Recruiter wars. Two recruiters call you within minutes of each other about the same role, then spend more energy fighting over who called first than actually finding you a job, and somehow one of them ends up angry at you for it. (This really happened to me).
- The non-compete trap. Recruiter-mediated contracts sometimes lock the candidate out of working directly with the client for a year after the contract ends. This does nothing for either the candidate or the client, it's purely to protect the recruiter's ongoing commission.
The fake job ad. Comes in three flavours: the same ad recycled every few months while a recruiter quietly harvests CVs for their database, the ad for a role that's already been informally filled but still has to be publicly advertised, and the outright scam ad whose entire goal is separating you from your cash.
Understanding Tech People
The appeal of a career in technology for people like me, is the endless supply of new puzzles to solve.
Job ads like the Hall of Fame give a whiff of "you must conform to exactly what we need" rather than "this is a place where we will value your contribution and you will learn and grow". A checklist like the one I received tells me that the recruiter is inundated by AI-generated slop CVs and is running a drastic slash-and-burn using AI, to narrow down the list and identify CVs with the most ticks, and this isn't really about people at all any more.
The rest of the Bad Behaviour section is just more of the same - recruiters and organisations optimising for their own efficiency, and leaving candidates dealing with the consequences.
Technology Should Make Things Better For People
My philosophy around technology has always been that it should make things better for the people or organisations using it. So it's reasonable to ask who's actually better off with AI-assisted recruitment, and I've come to the conclusion that:
- Job seekers aren't, since AI-written ads and AI-run screening have made it harder to get an application in front of a human.
- Companies aren't, since filtering this hard on tick-box criteria screens out capable people before anyone gets an opportunity to assess whether they're any good.
- Recruiters aren't, since they're flooded with applications which make it much harder for them to genuinely match the right person to the right role.
Looking Forward
Where does this leave us all? A few months ago I asked an AI tool to review my own CV, only to discover it was being identified as a beautifully optimised, circa-2010 CV. So I used AI to help me optimise it - for AI. I joked to a recruiter that AI writes the job ad, AI cross-posts it to every job board going, candidates use AI to mass-apply, and recruiter systems use AI to filter the applications. But where do the people fit in?!
Recently, a director of AI posted on LinkedIn about hiring for positions on his own team. Every resume he received was, in his words, "identical. Not similar. Identical." Flawlessly formatted and perfectly keyword-matched to the job description, to the point where he couldn't tell where the resume ended and the person began. He couldn't even use AI to sort the candidates, because when everyone feeds the same job ad into the same tools, the outputs come out looking much the same. His conclusion was that the most valuable signal left in an AI-run hiring process is evidence an actual human wrote it.
I left a comment under that post, which got more engagement than most of my own posts: "And the problem with not optimising for ATS is that you can't get through to an actual human in the first place. What a hopeless conundrum we've created. 😂"
I've got decades in the tech industry and I'm a strong candidate for a lot of roles, but all of these AI "enhancements" to recruitment only make me less likely to apply. I doubt I'm the only one.
The recruitment process is badly broken...
Regards,
A candidate your checklist just screened out, again.
