I've spent many years working in tech, and I've seen hiring change in a way that feels incremental if you're living through it, but transformative if you step back and consider the entire picture.
The hiring process used to feel very human in one way and very inconsistent in another. There was a resume to get you in the door. There was a recruiter call to check for basic fit. There was a technical screen to make sure you could solve problems under pressure. And then there were a series of interviews, typically involving engineers and managers, before a decision was made that sometimes felt very rigorous and sometimes felt surprisingly subjective.
That framework is still there. But what it's trying to measure has changed, and what it's trying to measure has created a different level of pressure on each step of the process.
The hiring process for tech jobs today is more process-oriented, more standardized, and more multi-layered than it has been in the past. It is no longer just designed to find people who are smart. It is designed to reduce noise, scale to meet large hiring needs, and collect enough data to make a decision in a world where both candidates and companies have more tools, more data, and more uncertainty than in the past.
From my perspective, what has changed most is that the hiring process is no longer just determining if a person can get a job done. It is determining if a company can rely on its data to prove to itself that a person can get a job done.
The résumé is still the first signal, but it is not enough anymore
A few years ago, the résumé had a different kind of power. A few key names, a good degree, some solid titles, and a good progression of events could shape the entire conversation. The résumé served as a kind of shorthand. A way to make assumptions.
Now, those assumptions aren’t as true.
Some of this is because of a more competitive market.
Some of this is because of a more savvy candidate.
And some of this is because of the power of AI to make even the least polished of candidates look good. A candidate can now create a respectable-looking résumé, a compelling summary, and a confident-looking application with less and less effort.
This doesn’t mean the résumé doesn’t matter.
This means the résumé is now simply an opening statement.
Relevance, depth, progression, and impact are still important.
But now, they are not enough.
Hiring managers want to make sure that the work they are seeing now will hold up to intense questioning later.
The uncomfortable truth: good candidates are falling between the cracks
This is the part of the equation that people in the industry don’t talk about loudly enough.
With the rise of technology and the increasingly automated and filtered process of finding candidates, good candidates are falling between the cracks before they ever get to have a real human conversation.
Sometimes it’s because the ATS doesn’t parse the resume well. Sometimes it’s because the profile doesn’t have the right keywords even if the person has done highly relevant work. Sometimes it’s because someone’s experience is great but just not presented in the expected way. And sometimes it’s because we prioritise speed and scale in our technology, even if it means missing some interesting but unconventional profiles.
This means that the process is not only becoming more strict. It is also becoming more specific.
A candidate might be qualified, seasoned, and prepared. They might still fail to get in because they failed to phrase their work in the exact words that the system was designed to emphasise. Another candidate might not come from the more trendy background, the more unconventional route, the smaller firm, or the job title that fits the more standardised model. They are not rejected or passed over because they were not seriously considered in the first place.
This is one of the not-so-obvious penalties of the more streamlined hiring process. Efficiency is gained, but some deserving candidates are lost in the space between the systems.
The recruiter screen is more important, not less
A great number of candidates are not taking the recruiter screen seriously enough. They think it is just a scheduling appointment, but it is one of the earliest tests of alignment and sincerity.
This is where the process starts to put the candidate’s story to the ultimate test. Why this job? Why now? What kind of work are they really doing? And how clearly can they explain themselves without resorting to jargon?
The more the front end of the hiring process is made more efficient, the more the recruiter screen is called upon to do the more nuanced work. A good application might get them in the door. A live conversation might reveal whether or not they are the real deal.
This is one of the less obvious effects of AI on hiring. “The easier it is to improve the written artifact, the more valuable the spoken explanation becomes. Not because the spoken explanation is inherently better, but because it is harder to pretend to be deep.”
However, there is an interesting tension at play. The recruiter screen is becoming an increasingly important part of hiring, while at the same time fewer candidates are passing through it. This means that the stages that occur before the recruiter screen now have significantly more power than they ought to.
The technical filter has become a stricter process because the funnel is larger
One of the biggest changes in the hiring process for tech people is the early stages of the technical filtering process.
It used to be that companies would invest engineering time early on. There used to be fewer stages between applying for a position and having a technical discussion. Nowadays, it seems that many hiring processes now put a formal technical screening process in place before the applicant is ever presented to an interview panel.
This makes sense from the hiring company’s perspective. There are so many applications, and engineering time is expensive. A quick filtering process is necessary to eliminate candidates before investing time interviewing them.
The technical screening process is now the front door, not the middle door.
However, the technical screening process is the area where AI has had the biggest effect on hiring.
The traditional assumption behind the technical screening process is that if you give an applicant a problem, the output will tell you about their skill level. This is no longer necessarily true. The applicant can now solve the problem differently because they have access to AI tools.
So, the technical screen is now in a strange, yet interesting place. The question still is, "Does the candidate solve the problem?" However, the question is also, "What does it mean to solve the problem?" Was it through deep understanding, familiarity, practice, tool usage, or judgment?
That, in turn, is affecting the way the technical screen is conducted.
Coding rounds are no longer just about code
To me, this is one of the most visible changes.
I recall a time when, in a technical screen, being able to write working code in a timely fashion was, in many ways, the ultimate question. It still is, to some degree. However, it is no longer the ultimate question in the way it used to be.
It is not the presence or absence of working code, in and of itself, that is as important today. It is what leads to the presence of working code. How does the candidate break down the problem? Do they clarify assumptions along the way? Do they communicate their thought process in a way that is understandable to others? Do they thoughtfully address edge cases?
These are all questions that, in some form, have always mattered. However, AI has pushed them to the forefront. When the cost of producing code is lower, the cost of producing judgment is higher. When answers are easier to produce, the interview starts to shift to whether or not the candidate actually understands the answer well enough to defend it, modify it, and apply it properly.
In many ways, technical screens today are no longer about coding; they are about engineering thought. And, to be honest, perhaps they should be.
Because, let’s face it, in real life, engineers are not paid to produce code. They are paid to make decisions.
System design is now an area where candidates can demonstrate their skill, especially outside the highest-level positions.
That makes sense, especially with today’s environment. Today’s engineering is rarely, if ever, an island unto itself. Even people who are individual contributors are still operating within some system, some constraints, some dependencies, or some trade-offs. They need people who can think about not just the function or the class, but about dependability, scalability, ownership, failure, maintainability, etc.
The effect of AI on this round is not as obvious, but it is still relevant.
If AI can assist with that, then architecture, decomposition, etc. become that much more significant. The real challenge shifts further up. The real challenge is, “What gets built? What does it look like? What does it look like when it’s put together? What are the risks? Where can I simplify?”
That’s why system design is now an important part of hiring. It’s an assessment of whether or not someone can think about structure when the world stops being so neat.
The results from behavioral rounds are now more significant than candidates realise.
Candidates often underestimate the importance of behavioral rounds. They often view behavioral rounds as easier or less significant than technical rounds. From what I can see, they are incorrect.
The results from technical rounds tell me if someone is capable. Behavioral rounds tell me if someone is trustworthy with that capability within a real-world team.
Can they work through ambiguity?
Can they disagree with others?
Can they own mistakes?
Can they make decisions without creating chaos around themselves?
Can they work with other people, other departments, other personalities?
These questions are more relevant now because modern engineering work is more collaborative, more interdependent, and more visible. Being a great individual performer is certainly good, but it is not sufficient.
It’s not just that AI has changed this round. AI has actually had a subtle impact on this round. This means people can practice their stories better, their examples better, and their presentation better. This means people are asked more questions. Specificity is required. Ownership is required. Depth is required.
A good story gets noticed. A good story that stands up well to further questions earns belief.
It’s a more layered process because trust is more difficult to earn. When people complain about the hiring process, they’re often right. The process is longer. The process is more repetitive. The process can feel like it’s been over-engineered.
It’s not because of the complexity of the process. The complexity of the process is a result of something much simpler. Trust. We want fewer wrong decisions. We want more than one indicator. More than one interviewer. More than one medium. More than one type of information.
And this instinct is only strengthened by the impact of AI.
If it’s easier to polish a resume, trust it less. If it’s easier to get coding help, scrutinise it more. If it’s easier to practice communication, ask more questions.
The result is a more layered process. Not because we like complexity, but because we don’t trust simplicity.
The irony is that it’s a more un-human process, even as it’s a response to a lack of trust in what’s real.
Where the process is letting deserving people down, this is where I think we need to have a more honest conversation.
The difficulty with hiring processes in general is not that they're difficult. The difficulty is that they're not necessarily blind in a discriminatory manner.
They're very good at handling large numbers of people. They're not necessarily good at handling non-standard strength in those people. Someone with significant experience but not necessarily good at writing a resume may not make it through. Someone returning to work after a break may not make it through in a position they're qualified for. Someone working at a smaller company may not make it through because their brand recognition is not as high. Someone with significant engineering sense may not make it through until it's too late and they're not selected in the final round.
The more we bring in AI and automation, the more this risk increases because we're getting much better at finding people we expect to find. We're not necessarily getting much better at finding people we do not necessarily expect but could be good hires.
This is why so many qualified people feel invisible in hiring processes. The issue is not necessarily that they're not qualified enough; it's that they're not necessarily seen in hiring processes.
What AI is really changing in tech hiring
A lot of people talk about AI in hiring processes as though it's just an efficiency play. And it is changing efficiencies in hiring processes. Recruiters are able to move faster, candidates are able to move faster, and resumes and other application processes can be refined faster than they were in the past. Interview processes can be scaled in a way that was not previously possible.
The thing is, though, this is not necessarily the real change brought on by AI in hiring processes.
The real change is that it's forcing hiring processes to think more carefully about what they're really measuring in each round.
If a candidate has used AI to improve their resume, what does their resume mean?
If a candidate has used AI to practice their answers to technical interview questions, what does their performance in an interview mean?
If a candidate has used AI to help them code, what does their skill mean?
If the work of engineers is increasingly being performed by AI, should interviews ban AI use completely, partially, or measure how well a person can use it?
These are not small questions. They get to the heart of hiring.
And yet, as a field, we are still in the middle of figuring them out.
The hiring process has changed. And it has changed recently.
It still looks like it used to. But beneath the surface, what it means to get a good signal has been up for negotiation.
What stands out to me most
I have been a part of this industry for a long time. And I don’t think the biggest change is that hiring has gotten harder.
I think the biggest change is that hiring has gotten more suspicious of easy signals.
A good resume is not enough.
A good phone call with a recruiter is not enough.
A correct coding solution is not enough.
A good behavioral response is not enough.
They want it to feel consistent across all of it. They want it to feel consistent with the story, the thinking, the technical depth, and the judgment.
And since AI makes it easier, they want it to feel consistent with that, too.
There is another story, however, that is playing out behind the scenes.
As companies become less trusting of the process, they become more dependent on tools that often fail to see the real talent. This means that not only are some candidates not being rejected on their lack of merit; they are not being rejected at all. They are being eliminated prematurely, mechanically, and sometimes incorrectly.
This is the trade-off that the industry still has not figured out.
The process, stripped down
Despite all the changes, the process still looks vaguely familiar.
You apply.
You screen.
You pass the first technical screen.
You proceed through multiple rounds.
Interviewers confer.
A hiring decision is made.
This is still the process.
What is different is how they interpret each step along the way.
The resume is now considered a hypothesis.
The screening process is now considered an automation process, at least in part.
The first technical screen is now considered a calibration process, at least in part.
The coding process is now considered a reasoning process, at least in part.
The behavioral process is now considered a trust process, at least in part.
The hiring process is now considered an integration process, at least in part.
This is what hiring looks like today.
It is not broken.
It is not necessarily better.
It is, however, different.
And the role of AI is not separate from the hiring process. It is not an adjunct. It is not an add-on. It is not an afterthought. It is an integral part. It is an integral part of what hiring means.
It is also changing who is seen, who is filtered out, and who does not even get a chance to prove their capabilities.
