When I think about hiring, I don’t think about roles anymore.

I think about skills. Who on my team can play multiple roles using AI. What specific capability are we missing that we can’t build ourselves.

This isn’t theory. It’s how I’m building my product team right now.

What I have seen this year

My Product Manager is doing AI Engineer work. Junior Product Analyst is doing Data Analysis and User Research. They are building prototypes using off-the-shelf SaaS tools and setting up infrastructure locally. Work that would have required a specialized hire an year or two ago.

Recently, they validated documentation in a fraction of the time it would have taken with a dedicated team. The business impact isn’t just speed. It’s the ability to test quickly. Build fast, break fast.

Here’s what I’ve learned: if you already know exactly what needs to be built and you need to build at scale, you still need a specialist. But if you’re not sure and want quick prototypes and testing, AI lets you reduce your team and headcount dramatically.

Prototype Teams vs. Scale Teams

As PartRunner grows 3x or 5x, I’m determined not to make one mistake: hiring a lot of engineers to build something that scales before we know what works.

A lot of resources go into “understanding” what works. That’s the expensive part most companies get backwards.

I’m building a permanent structure with two distinct types of teams:

Prototype teams will be nimble and agile. They’ll integrate into any team in the company and run quick experiments. These aren’t temporary roles. They’re a permanent capability. I would request speed first and scale second

Scale teams will work with prototyping teams to see how to scale what actually works. They come later, when you know what you’re building. I would request scale first and speed second

The Job Titles Consolidation That Are Coming

In 2025-2026, we will see more consolidation of different Job Titles

AI Product Engineer — someone who combines product thinking with the ability to build and prototype using AI tools. They don’t just spec features. They build them.

Product & Technology Manager — a hybrid role that doesn’t hand off between product and engineering. They own both.

These aren’t speculative. AI-related roles like AI Developer, Prompt Engineer, and Machine Learning Engineer are already among the fastest-growing in the U.S. Typical AI development teams now include just three to five people: a Product Strategist, an Engineer, and a QA lead. Compared to traditional dev teams with 8-12 people and multiple layers of handoff, these lean teams deliver more with less. AI-assisted software development speeds delivery by up to 45% while reducing costs by 55%.

The Two Skills That Matter Now

If you’re on a product team today, here’s what you need to learn in the next 6 months:

First: If you’re good at something, use AI to become 2-3x faster at it.

Second: If you don’t know something, use AI to become familiar enough to have the basics down.

This isn’t about replacing expertise. It’s about expanding your range. The people who can do this will be the ones companies fight to hire.

McKinsey research shows that companies blending AI with human decision-making are 20% more successful in rolling out AI projects. The key word is “blending.” You’re not becoming an AI. You’re becoming someone who knows how to use AI to do work that used to require three different specialists.

What This Means for Your Hiring in 2026

Stop hiring for roles. Start hiring for skills.

Look at your team and ask: who can play multiple roles with AI? What specific capability are you missing that you genuinely can’t build internally?

If you’re in the experimentation phase, you don’t need specialists yet. You need people who can move fast, learn quickly, and aren’t afraid to break things.

When you do hire specialists, you’ll know exactly what you’re asking them to build. You’ll request scale first, not speed. That’s the difference between wasting money on premature optimization and investing in capabilities that actually matter.

The companies that figure this out will build better products with smaller teams. The ones that don’t will keep hiring the way they always have and wonder why their AI initiatives aren’t delivering returns.

I’m betting on the former. At PartRunner, we’re proving it works.