Andrew Ng was right: the jobpocalypse is canceled. We keep hiring, and we keep looking for new talent. With AI we test more formats, we produce more content, we run more platforms than we could a year ago. And more output needs more people to own it.
My hiring filter has changed a lot this year, and I wanted to check it against people who hire at scale. So I asked 8 founders and CEOs the same question: who are you hiring right now, and what did that person do to get in front of you?
Their companies range from a 170-person team to thousands of employees. Not one named the resume as the thing that decides it. Not a degree, and not even years of experience. Every single one described work they had already seen before the interview started.
They google you first
Sal Khan, founder of Khan Academy, runs a team of 350 people. He explained it best.
Imagine this, someone today graduates from Stanford, pick your major, good major, with a 3.9 GPA. When I look at that resume, I say, okay, probably smart, got into Stanford, good GPA. I don't know if those skills they learned are gonna be directly transferable.
What I normally do now when I see a resume, even someone who's been working, no matter what they write, I look for YouTube videos of them.
He wants to see how a person communicates and thinks before he ever meets them.
We do the same thing. My team checks the candidate's LinkedIn: what they post and the way they think. If you're visible online, you're already one step ahead of most candidates.
A strong LinkedIn profile is a must if you want to stand out.
Explorers, not map readers
Yamini Rangan runs HubSpot. She said she hires for three things now:
A scientist mindset. Hypothesis, test, learn if you were right or wrong. She asks for examples of when a candidate actually did this.
Depth in the work. Knowing which parts of a workflow are broken and which ones AI can improve.
Curiosity about the customer. Using AI to solve something for a customer, not for its own sake.
The old playbooks were maps. Someone figured out the path, you followed it.
Today, with AI, there is no map. So you have to get comfortable with being an explorer.
Yamini Rangan, CEO of HubSpot, on the podcast. She shared a lot of advice on how to get noticed and hired in 2026.
Luana Lopes Lara co-founded Kalshi, a 170-person team. The "how have you used AI for this?" question started in their engineering interviews. Now it's in the design interviews, and the legal team is asking it too. She said it's less about having the answers and more about whether the person is willing to try.
When we were hiring a social media manager this year, every question I asked was some version of the same one. Where does AI go into this process? I don't need the right answer. The answer I want to hear: "I'm not sure, but here's how I'd test it, and here are the two things I'd try first." That's what I'm hiring for now, not just past experience.
One level deeper than the job requires
Aaron Levie has run Box for 20 years and hires across marketing, sales, finance, engineering. I asked how technical a candidate needs to be. His line was specific:
You don't have to be vibe coding all the time and building entire products, but you should try and really understand what is the agent doing? How does it work? How does MCP work?
Domain skills still matter to him: knowing customers, knowing how to sell. AI sits on top of that. His starter advice for non-technical people: download Codex and Claude, give them a real problem, wire them to your own data, and watch what happens.
Everyone on the team builds
Andrew Ng has taught more than 8 million people AI, so I asked him for a benchmark: what does "good with AI" look like when he interviews a marketer, not an engineer?
All of my marketers know how to code. So as part of how I interview marketers, we ask them what they've built.
His example: someone on his marketing team built a desktop app that crawls the web for related articles before he writes, then lets him chat with the whole thing. His CFO built scripts that open files, check them for consistency, and flag what needs attention, instead of her team clicking through documents for hours every week. Nobody waited for an engineer.
He applies the same rule to every role: marketers, recruiters, HR, operations.
If they learn to build with AI, they'll really just do much more, whatever the job role is.
Grant Lee, co-founder of Gamma, sees the same pattern on his team. His head of design was writing code before vibe coding made it easy for designers, and now takes a prototype from first sketch to production himself.
My YouTube producers built a guest ranking system in Claude that estimates a potential guest's audience and how a video with them could perform, before we reach out. It used to take hours: researching their topics, checking how visible they are across other shows and platforms, finding a news hook, and sketching an angle for the interview. Now Claude does all of that in 2 to 3 minutes. And no, we don't rely on it 100%, but it's right more often than not. Two years ago I wouldn't have known how to build something like this, with an engineer or without one.
One guest scored in our Claude ranking system: 38/40, Tier 1, with a forecast range for the video.
What was essential a few years ago has already shifted. Ryan Roslansky from LinkedIn confirmed it with data: the skills required for a given role changed by more than 25% in the last couple of years. His forecast is that the shift reaches 70% by 2030.
🔴 The resume is still the gate
And yes, I got a lot of comments under the video saying the same thing: a machine reads my resume before any of these people do. True.
So the resume stays. It gets you into the room. What you do in the room is what changed. Conor Grennan, CEO of AI Mindset, put it best:
Go in there and say: here's my domain expertise. But let me just show you what I would do with AI and how I would reinvent this process.
His point: anyone can say "I know how to use AI." Walking in with a plan for how you'd rebuild the role is what nobody else in the pile does. That's where the interview time goes now, not into defending the resume.
I put the full conversations into one video.
One takeaway I want you to leave with today: invest your time in learning and building with AI. Stay curious, and get comfortable with change. What you build is your demo. Then write about it where the people who hire can find it.
Marina 💜
P.S. Want more on this? Here are the 7 skills that came up again and again in these conversations — read here.
