Your Resume Can't Capture the Skill That Matters Most Right Now

"This might not make any sense, but stick with me here..."

I've said that more times than I can count. Mid-workshop, mid-meeting, mid-conversation. Something would pop into my head. It's always an image for me. A screen, a flow, a metaphor in visual form. And I'd hedge with that line before trying to lay out what I was seeing.

Sometimes it landed and the room lit up. Sometimes it was blank stares and crickets. Sometimes I'd realize mid-sentence that what I was saying made no sense.

The picture in my head was always clear. The words just couldn't carry it.

I'd never designed a product before. I'd had opinions about them. A lot of opinions. Constantly. Some people would call that a character flaw. But it's always helped me move faster because I don't just have the opinion. I think about it until I understand why I have it. For most of my career, though, that's where it stopped. Hoping someone else could see what I was seeing.

With my co-founder Nilima, that started to change. I could explain where my brain was going and then go prototype something to help her see what I meant. It wasn't always perfect. But it got us a hell of a lot closer than my verbal explanation ever could.

And then it kept going. I kept finding myself doing more and more things I thought I'd never be able to do. Setting up tracking for our product. Figuring out how to measure what users were actually doing. Building our Stripe checkout pages. (Did Nilima have to fix some of it? Yes. Did I build the first version myself? Also yes.) Things that, a year ago, I would have said were firmly outside my lane. I'm "non-technical," though Nilima hates when I say that (she wrote a whole post about why). Every time I turned around to look for the wall that was supposed to stop me, it wasn't there.

At some point I realized this thing that kept happening was a skill. One I'd never seen anyone name. And when I thought about what to call it, it felt obvious (like things usually do. It's annoying.)

Operational dexterity: a working definition

The way I'd explain it: it's the ability to take on something you've never done before, figure it out with whatever's available to you, and be OK with getting it wrong along the way.

It's a skill. But it's supercharged by a mindset. And it's not speed. It's the ability to pick up an unfamiliar tool and make something precise with it.

Some people do this naturally. They see a problem and their first instinct is to start building toward a solution, even if they've never done anything like it before. Other people will find this deeply uncomfortable. The ambiguity. The lack of a playbook. The possibility that you'll spend a bunch of time on something and it won't work.

But it can be developed. And right now, it matters more than almost anything else on your resume.

Good luck putting this in a bullet point

Resumes couldn't capture what mattered before. Now it's not even close.

AI didn't give people new ideas. People have always had ideas. What it did is collapse the distance between "I see a problem" and "I can do something about it."

You used to need a business case. A budget. The right title and a team. And you used to need permission. Now you can identify a problem, prototype a solution, and in some cases, actually ship it. But even if you never ship anything, you can build enough of the thing to get other people to understand the problem quickly. You can make the abstract concrete. That alone is a massive shift.

The barrier between seeing something and doing something about it has gotten dramatically lower. Operational dexterity is what happens when someone actually walks through that door.

Now hiring: Builders

This isn't about being good at AI tools. I want to be clear about that.

This is about being a builder. Someone who sees a gap and starts filling it, even without a roadmap. The tools just mean you can act on that instinct faster and with less dependency on other people's timelines.

Nikhyl Singhal (former product exec at Meta and Google) said something on Lenny's Podcast recently that I haven't been able to stop thinking about. He argued that the product management role is splitting in two: "builders" who ship and prototype and have direct product instincts, and "information movers" whose job is mostly framing and routing information up and down the org. His prediction? Companies will shed tens of thousands of people and rehire thousands. All AI-first. All builders.

He was talking about PMs. But that split isn't a PM problem. It's a knowledge-work problem. Every function has builders and information movers. And AI is eating the second group's job description first.

Here's the catch, though. The tools are largely being built by the people who already know how to do the thing. So they're optimizing for themselves without realizing it. Vibe coding is a perfect example. It's mostly being developed by engineers. Which makes sense. But they're missing a huge demographic of people who could build things if the on-ramp didn't feel like a cliff. An entire market segment, just sitting there.

I've been shamelessly reaching out to the Claude Code product team about this. The people who shy away from anything remotely technical, especially women and female founders, are right there. And nobody's building for them yet.

So yeah, there's a mountain. These tools weren't built for us. The language is unfamiliar, the interface assumes you already know things you don't, and there are moments where you genuinely have no idea if clicking "yes" is going to break everything. I know because I've been there.

But the other side of that mountain? It's wide open. And the people willing to climb it, even awkwardly, even slowly, are going to have access to things that weren't possible two years ago.

Generalists, to the front please

I'm a generalist. I've always been a generalist. And this is our time to shine.

When some people look at something they've never done, they see a wall. When we look at it, we see something we just haven't figured out yet. By trade, we've worked across functions and had to learn new things constantly. Out of necessity (and curiosity) we stopped assuming we couldn't and started believing we'd figure it out.

You don't need to master something forever. You need to master it enough for the thing you're building right now.

That mindset has always been undervalued. But now the tools actually let us act on all of that cross-pollinated thinking. The generalist who's been in strategy and operations and marketing and has strong opinions about product? They can now prototype those opinions. They can build things that show how their different perspectives connect.

I think having different perspectives coming in through more avenues will help the business landscape evolve in a more human way. Not just faster. More human. Because the people entering these conversations now aren't all coming from the same background or the same set of assumptions. And often, the people with the most diverse experience are the ones without the big logos on their resumes. Ironically, the environments with the most cachet and superficial signal are often the ones that train these muscles the least. But that's a whole other post.

I have opinions, not a crystal ball

I don't know what happens when everyone develops this skill. Right now, operational dexterity is a differentiator. But if the tools keep getting better and more people build this muscle, what's the next layer? I don't have that answer, either. But I know this: operational dexterity isn't something you put on a resume. It's something you demonstrate. So go build something. Even if it's messy. Even if someone has to fix it after you. The fact that you built it at all is the point.

I linked some resources below to help you get started.

Because at its core, this is about more people being able to show what they see. And I think humans building things is the point. It always has been.

The tools changed. The instinct didn't.

Start building

• Theanna — community for women building in tech
• Claude Code — for when you're ready to climb the mountain
• Claude Cowork — for automating the work around your work, no coding required
• Lovable — describe what you want, get a working app
• Nikhyl Singhal on Lenny's Podcast — the builders vs. information movers episode
• Humans Build Things. That's Not Going Away. — the companion piece
• Technical (adj.): Finally Got With the Times — Nilima on what "technical" actually means