Posted by In 2006, Amazon did something that had nothing to do with selling books. They took the infrastructure they’d built to run their own systems and opened it up to the world. Suddenly, a two-person startup in a garage could access the same computing power as a Fortune 500 company. The cost of building software dropped by an order of magnitude. The barrier to entry collapsed.
AWS didn’t improve infrastructure. It reinvented who could have it.
That’s what an AWS moment looks like. Not a better version of what existed before. A fundamental shift in what’s possible and who it’s possible for. Once it happens, going back is unthinkable.
Technology has had several of these.
Netflix didn’t improve video rental. It made the category irrelevant. Android didn’t make phones cheaper. It made computing mobile for two billion people who couldn’t afford what came before. Real-time payments didn’t speed up bank transfers. They changed what money could do and who could access it.
Each of these moments followed the same pattern. A category that seemed complex, entrenched, and resistant to change suddenly became something else entirely.
Now think about hiring.
In the same twenty years that produced AWS, Netflix, Android, and real-time payments, what happened in recruitment technology?
We got better sourcing databases. Faster applicant tracking systems. Resume parsers that actually work. Interview scheduling tools. Video interview platforms. A lot of workflow software that made the process of hiring smoother.
None of it changed the fundamental question every hiring decision comes down to: can this person actually do the job?
The closest thing hiring got to an AWS moment was the resume going digital. That happened thirty years ago. Since then, the category has iterated. It hasn’t transformed.
There’s a specific reason. AWS had a clean, objective output: speed, uptime, cost per compute unit. Hiring’s core output is a judgment about a human being. That’s harder to standardise, harder to measure, and deeply uncomfortable to hand to a system. So the industry optimised what was easy to productise and left evaluation, the thing that actually determines whether you hire the right people, largely untouched.
The consequence of that choice falls unevenly.
It falls hardest on a specific kind of candidate. The one with real skill but without the referral network. Without the educational background that reads as a signal before a single question is asked. Without the profile that feels low-risk to a panel making a quick call between meetings. The system wasn’t designed to exclude them. It just wasn’t designed to find them either.
Before GenAI made evaluation fashionable, we manually reviewed more than five million lines of code to build our own assessment systems. We kept finding the same thing: people the existing tools couldn’t see properly.
One of them had been selling coffee for three years after graduation. Engineering interviews kept going nowhere, so he found other work. When he came through our system, evaluated on what he could actually demonstrate technically, he cleared the bar. He eventually landed a role at an AI startup.
He didn’t need a different kind of opportunity. He needed a system that could actually see him.
That’s the part of the AWS story that doesn’t get told enough. The compute was never the point. The point was that access stopped being rationed by who could afford the old infrastructure. A two-person team had what the Fortune 500 had. The quality of your idea became the variable. Not the size of your balance sheet.
Hiring still rations access. Not through cost, but through inconsistency in evaluation. And the people who absorb that inconsistency most are the ones without the backup systems. The referral. The brand-name employer on their CV. The connection who flags them as worth a closer look.
We’ve placed more than 2,000 developers at companies like Zomato, Intuit, Target, and Siemens. Some would have found their way through the existing system regardless. But a meaningful number wouldn’t have. An operations professional who went on to work on Google devices. Developers who had been in the pipeline for years, skilled, but invisible to a system that never gave them clean signal to stand on.
AI changes this, but not by replacing interviews. By making evaluation consistent enough that the outcome reflects the candidate, not the circumstances. For the first time, the infrastructure of assessment doesn’t have to depend on who shows up to conduct it.
Hiring’s AWS moment isn’t going to look like a better interview. But a fairer one.