ATRIUMsearch → argument graph
ClaimVideo · 55:19 — 57:25

The most valuable hiring criterion for growth roles is systems thinking — the ability to decompose your job into inputs, outputs, and automatable components — because AI-native generalists who can build self-improving workflows will outcompete deep specialists.

Matt argues that hyper-specialists are dying and that the best growth hires are AI-native systems thinkers who can map their entire job as a system, identify automatable parts, and build self-improving agentic workflows — noting that less than 1% of candidates meet this bar. ✦ AI generated

Matt Swulinski · 20VC · 2026-08-15 · original ↗

starts at this moment · 55:19

Elicited by

How has what you look for in talent changed in the last year or two?

Now, I'm looking for that. But they don't necessarily have to be the best in the world historically. If they're AI native or a systems thinker and can deconstruct what makes their job hum, that person plus experience will out compete someone that just has experience and is not an AI native... to me that's being able to step back from let's say a task that is a part of my job being able to take one degree of separation from the task itself and say in my role what are all of the moving pieces that essentially I need to do on a day-to-day basis what are the inter relationships, where's the boring admin, where's the reporting, what's all of that. And being able to map that out and understand how your job as a system actually works, because then you can actually say, okay, I can apply an agent or AI to this part and now I can focus on this higher leverage part of my job... if you ask a lot of people that, it turns out they don't really understand their job.

verbatim transcript · starts at 55:19

Transcript · around this moment

54:59in let's say meta ads. They've scaled to hundreds of millions of dollars and they're, let's say, the best in the world as a meta ads media buyer. And that would be the main thing I'd be looking for. Let's say that's the person I'm hiring. Now, I'm looking for that. But they don't necessarily have to be the best in the world historically. If they're AI native or a systems thinker

55:21and can deconstruct what makes their job hum, that person plus experience will out compete someone that just has experience and is not an AI native. And it's like I'm seeing that happen where like I'm going to have a hot take here. I think um a lot of teams are trying to make their teams AI native and they're failing. >> Why are they failing? the people that are in those roles are

55:44not systems thinkers where like if you're in a marketing or let's say ops role and you're not a systems thinker you don't know you actually don't know >> how do you think about like what does a systems thinker mean >> so to me that's being able to step back from let's say a task that is a part of my job being able to take one degree of

56:03separation from the task itself and say in my role what are all of the moving pieces that essentially I need to do on a day-to-day basis what are the inter relationships, where's the boring admin, where's the reporting, what's all of that. And being able to map that out and understand how your job as a system actually works, because then you can actually say, okay, I can apply an agent

56:23or AI to this part and now I can focus on this higher leverage part of my job, right? And a lot of people go and go to chat GPT and ask it a question and then they go on still doing their job the manual way, right? And if you take a step back and say, "What would it take for me to fully automate, let's say, all of my job, and of course it's not going

56:45to work. Things are going to break apart." But run yourself through that thought experiment. And the thing is is if you ask a lot of people that, it turns out they don't really understand their job. And because they're not systems thinkers, they they don't actually know what are all the inputs and the outputs. Uh so that that's the line for me. So like when I when I

56:59interview people to fill in the team, uh it's it's like partially like an engineering interview where like systems thinking is the bar. >> So what do you ask them to do? Well, I I want a job with you. Okay. You're clearly [ __ ] brilliant and I'm a podcaster. Um what are you going to put me through? How are you going to test me? >> Yeah. So like core question is more of

57:20like how do you use AI as a workflow in work? But I'm actually more curious, how do you use AI workflows in your personal life? Like do you have any systems that you feed through AI that make you a better better person outside of work? And like maybe the line blurs of like what you do is also work is personal whatever. >> And so a good answer versus a bad answer

Related moments