A company called Andela just looked inside tens of thousands of tech job listings and found something messy: half the people writing "AI Engineer" or "ML Engineer" on a job posting don't actually know what that job should include anymore.
Andela pulled data from over 47,000 technical job postings at Fortune 500 companies and scored more than 2,000 specific skills mentioned inside them. Among postings titled things like AI Engineer or ML Engineer, more than half asked for skills that belong to two different, older jobs mashed together. A machine learning engineer today is often expected to also know software architecture. A DevOps person is expected to know cloud engineering too. Front-end developers are being pushed to think like product managers.
This is not random confusion. Andela's head of research, Cory Hymel, calls it "skill bleed": when a new type of job is forming, pieces of it leak out of old job titles before anyone agrees on a new name for the whole thing. His team ended up naming eight of these new blended jobs themselves, including FinOps reliability engineer, someone who tracks cloud computing costs and the cost of running AI models at the same time, and LLM application engineer, someone who builds products on top of existing AI models instead of training new ones from scratch.
This kind of merging is not new in tech. DevOps was born years ago when software developers and the people running servers got tired of working in separate silos. Security got folded in later, creating DevSecOps. What is different now is speed. AI is changing what a day of work looks like every few months, not every few years, and job descriptions cannot keep up.
There is a second, harsher story running alongside this one. While some tech jobs are blending into bigger, better-paid roles, other tech jobs are simply vanishing. Entry-level coding work is shrinking because AI now does much of what junior developers used to do, and some large tech employers have openly said they are not hiring new graduates this year.
So two things are true at once: AI is destroying some narrow, repetitive tech jobs, and it is fusing others into broader roles that pay more but demand a wider mix of skills. The mistake most companies make is treating this as one problem with one fix, like a hiring freeze or a round of layoffs. Hymel's advice, backed by his own data, is the opposite: companies are actually short-staffed in these new blended roles, they just don't know how to describe them yet.
The lesson is not really about tech hiring. It is about what happens to any job once AI takes over the narrow, repetitive part of the work. The boundaries of the role shift, the job title stops matching the job, and whoever writes the job description gets there last. That is already happening in tech. It will not stay there.