Elton Chan is the Co-Founder of Second Talent, a solution that connects global tech leaders with AI-native engineering talent across Asia.
I’ve become a little suspicious of the idea that AI automatically means fewer hires.
A few months ago, one of our clients was planning to hire eight engineers. They introduced AI into the development process, found they could get the same work done with five and cut the requisition. Then those five started getting through work faster than expected. By June, the company was asking us to hire 11.
It sounds contradictory until you look at what the company did with the capacity it created. They started taking on work that had previously been too expensive to justify, and that created demand for more people.
I’ve seen versions of this in several parts of our hiring work.
1. Freed capacity gets spent, usually within two quarters.
I’ve stopped believing companies simply bank the efficiency.
When a team gets more productive, the conversation usually moves to the work they couldn’t justify before. Maybe there was a market they wanted to enter but couldn’t staff properly, or a product that kept getting pushed back because engineering was already at capacity. Sometimes it’s a customer segment where implementation costs made the economics difficult.
We’ve seen this with clients after their engineering teams became more productive. Work that had been sitting on the roadmap gets picked up, customers who weren’t previously viable become worth pursuing and that creates demand for people outside engineering, too.
A company might need fewer engineers to maintain what it already has and still end up hiring more people because it can now do things it couldn’t afford to do before.
2. The demand has moved into the business as well as the technology.
The technical candidates we work with are increasingly bringing domain knowledge into AI projects, and that’s becoming more valuable as companies put AI into existing parts of the business.
The Anthropic Staff Software Engineer, GTM AI Engineering role is a good example. They’re hiring an engineer to build agents for go-to-market work while working with sellers, RevOps and platform engineering.
The engineering experience is still the foundation of the role. The difference is that the engineer is expected to understand the environment where the system will operate.
I’ve seen this come up in our own searches. An engineer who has spent years building systems inside a particular business function can bring context that another technically strong engineer may have to acquire after joining.
That changes how I look at technical candidates. I’m still looking closely at the engineering work they’ve done, but I also want to understand the environments they’ve worked in and the problems they’ve solved there.
3. AI is moving out of the AI team.
A few years ago, it was easier to think of AI as something that belonged to a dedicated machine learning or data science team. That’s getting harder to do.
Lightcast’s 2025 AI labor-market research found that more than half of job postings requesting AI skills were outside IT and computer science in 2024. Its latest analysis also found that mentions of agentic AI skills in U.S. job postings grew by more than 280% in a year.
That lines up with what we’re seeing from companies we work with. They aren’t necessarily creating a new AI department. They’re adding AI to systems and workflows that already exist, which means the technical work starts spreading across the organization.
A company can have strong engineers who know the technology and still need people who understand how that technology has to operate inside the business.
The AI capability might sit in engineering, but the person using it could be in sales, operations, finance or customer support.
4. Entry-level work has moved up.
I expected AI to reduce the number of junior roles we filled. Instead, the bigger change has been in what companies expect from those hires.
Two years ago, a decent share of the junior roles we filled in Manila and Cebu involved regression testing, catalog data cleanup and building product pages against a specification somebody senior had already written. A new employee could learn the systems, get familiar with the product and gradually take on more complicated work.
A lot of that first layer is now handled with AI.
The junior specifications coming to us increasingly ask for someone who can investigate a failing pipeline or decide what needs to be escalated. That’s a very different starting point for someone early in their career.
It also creates a problem companies haven’t fully solved. Those repetitive first jobs were how people learned the domain. Someone spent a year or two fixing small problems and working under more experienced engineers before they were trusted with something ambiguous.
If AI handles much of that work, companies have to think more deliberately about how a junior engineer gets that experience.
Some of our clients are giving junior employees exposure to live support and production work earlier. Others are pairing them more closely with senior engineers. And all this is because companies can’t assume the old progression will happen on its own.
AI doesn’t automatically mean fewer people.
I don’t think we can say yet whether AI will create more jobs than it removes. There are too many differences between industries and companies to make that prediction useful.
What I can say from our hiring work is more specific. Companies are using the capacity created by AI to take on work they previously couldn’t justify. They’re also looking for people who understand the business around the technology, while junior roles are starting further up the ladder than they used to.
AI is changing what companies can get done with the people they already have. In a lot of cases, that is giving them a reason to hire for what comes next.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?