With the rise of AI assisted development, I'm seeing a new trend in tech hiring.
More and more companies are looking for a "generalist" [sic] either explicitly or under the fancy label of product engineer.
For the last 20 years, the tech industry has been looking for specialists, engineers who mastered specific languages, frameworks, vendors or roles.
Meanwhile, a smaller group kept jumping from role to role, language to language, using whichever framework did the job (or was leftover from the previous team), or adopting vendor tools by mandate; all in pursuit of a higher-level goal: solving real-world problems they had "fallen in love" with.
While job hoppers chased the next big deal, trained for technical interviews and maximized their narrow specialization, generalists tended to stay longer. Generalists got invested in the problem and developed the broad skills needed to clean up the brownfield mess left behind.
Apart from that, the cost of job hopping for a generalist was too high: every move meant starting from zero and the return of that investment on developing new skills takes at least 3 to 5 years if not more.
Hiring is hard and tech startups (and those who fund them) see AI as the solution. Specialists are no longer the priority, the search has shifted to those willing to join the pieces together.
Unfortunately, you cannot evaluate generalists through a skills based process. They will give average answers to most problems, often struggle with live coding interviews and may find system design challenges awkward.
What matters is harder to measure: how they figure things out, their agency, long term execution, and willingness to take on the unsexy work. That doesn't show up in an 8 step mechanical interview process.
I'm curious to see how this "new role" evolves and how the industry adapts to actually find them. In the end, the real test isn't for generalists, it's for the companies trying to hire them.