Wade Foster, co-founder and CEO of Zapier, returns to Beyond the Prompt to talk about what it means to refound a company after 15 years. He joins Henrik and Jeremy to discuss why AI is pushing him to question old ways of working, how to get people excited about change, and why the ability to act without having all the answers matters more than ever.
After 15 years of building Zapier, Wade Foster is inviting his team to rethink almost everything. What started as a “code red” response to AI has become an opportunity to rediscover what the company could be. Wade describes the feeling as “10 out of 10 anxiety, but 10 out of 10 fun.” The challenge is getting an entire organization to see the possibilities, especially when change can feel threatening.
Wade shares what this looks like inside Zapier, including an experiment to let AI agents handle customer support emails. It turns out that handing work over to AI is much harder than simply adding an assistant to an existing process. The team has spent months figuring out how to make it work, with humans still playing an important role.
Henrik, Jeremy, and Wade also explore what it takes to thrive when the old rules no longer apply. They discuss curiosity, resourcefulness, and the instinct to act instead of waiting for instructions. And while Wade is giving his team permission to question everything, the conversation raises a harder question: how do you get people to challenge the way things have always been done when that's exactly what they've been taught to follow?
Key Takeaways
Wade's LinkedIn: LinkedIn.com/wadefoster
Wade's X: X.com/wadefoster
Zapier: Zapier.com
Wade's articles: Zapier.com/wadefoster
00:00 Intro: The Work Is Changing
00:48 Meet Wade Foster
04:49 Is Your Strategy Too Small for AI?
06:14 From Code Red to Refounding Zapier
07:31 10/10 Anxiety, 10/10 Fun
08:50 How Much Has AI Actually Changed?
10:41 Getting People Excited About AI
13:56 Why Jobs Are Changing
15:34 Rethinking How Work Gets Done
16:39 Inside Zapier's AI Support Experiment
20:14 When Existing Workflows Break
21:45 Identity, Curiosity, and Adaptability
25:36 The Instinct to Act Without All the Answers
29:39 Stop Waiting for an Assignment
33:28 You Can Just Build Something
34:43 What Does Refounding Actually Mean?
36:40 Giving People Direction Without All the Rules
38:36 The Future of Automation at Zapier
39:18 Wade's Advice for Refounding a Company
41:45 What Would a New CEO Do?
43:23 The Debrief
[00:00:00] Wade Foster: Like we need more engineers than ever, but if the thing you fell in love with about engineering was like this artisanal handwritten code portion of the job, and that's the thing that you yearn for, I gotta be honest with you, like I do think that's a bit of a relic. And, you know, if the new job is about trying to understand customers' problems and translating that into systems that go build and architect software on your behalf, and you don't like that version of the job, then okay, you have a personal choice to make.
Hey, I'm Wade Foster. I'm the co-founder, CEO of Zapier. We are just celebrating our 15th year as being a company, and I just shared a memo with the organization on re-founding Zapier. Excited to talk to y'all about what it means to re-found.
[00:00:48] Jeremy Utley: Simon Sinek. He said, uh, I'd be curious to know what y'all think of this. I'm sure you resonate. He said, "The difference between..." And by the way, for the record, I don't agree with Simon Sinek on a lot of his AI takes, but, but I do think this is interesting.
"The difference between a startup and an enterprise is an enterprise's ambitions lie within reach, and a startup's ambitions exceed its grasp." And to me, it's a beautiful kind of articulation about the importance of ambition, and it strikes me this has never been truer than now. Like enterprises, all of the goals lie within reach.
They aren't thinking of something that's, that's beyond grasp. And- That requires, I don't know, spaciousness. I mean, this is something Henrik and I talked about with someone last week, spaciousness. Um, but Wade how do you think about sparking imagination, ratcheting up ambition of the organization? And maybe even just start with yourself.
[00:01:42] Wade Foster: Well, I mean, for myself, I gotta be honest, I think it's a learned behavior. Like, I don't think I came out of the womb, you know, thinking, "Oh, let's go to Mars."
Well, may- I, you know, may- maybe I did. Like, I-- But I do think the school system kind of beats that out
[00:01:59] Jeremy Utley: of you. I didn't. I d- for the record, I
[00:02:01] Wade Foster: did not think that. Well, well, so I'm, I'm thinking back to my childhood. It's fun 'cause like I, my, I have a f- four-year-old daughter and she's obsessed with space.
She's like reading all the space books and stuff, and it makes me happy because that was me as a kid. Like, I was... You know, I, I remember being in elementary school and poring over the school library reading every space book, and I m- I remember in each page, all I wanted to hear was the author say, "They're aliens."
Like, "They're aliens and you're gonna get to meet them." Like, that was the thing that like... You know, then you sort of go through the public school system and all that sort of stuff and like, I, I grew up in a conservative town. Like, a lot of that ambition kinda just gets beaten out of you. It's the, you know, they're, they're sort of like, "These people are the rocket science.
These people are the, like, professional athletes. These are the, you know, uh, incredible musicians." And like, you know, that kinda doesn't happen where we're... You know, you, y- you know, let's stick with what would be good for you. Like, maybe a doctor. Like, doctor is sort of like the, "You could do that." Mm-hmm. Or like a teacher- That's a safe bet
or like- Yeah ... those are, those are things you could do, right? You know, very, very much in, in line with, like, what an ambitious parent wants. It's like, you know, go, go do one of those respectable careers, right?
[00:02:59] Jeremy Utley: Mm-hmm.
[00:02:59] Wade Foster: So that's like, as I was kinda going through, a lot of that got beaten out of me, and where I relearned it, honestly, was in the act of doing Zapier.
I was fortunate enough to have two co-founders who had ambition and, you know, I think we kinda just like, you kinda egg each other on, you encourage each other. But even the back of your head, you still have your doubts, you still have your, like, question marks and like, "I, uh, is this gonna work? Is this gonna work?"
But, you know, we're now-- This month, Zapier turns 15. So we've been at it for 15 years now. And so for me, I've experienced so many moments where we have achieved something that when I look back to Zapier 15 years ago, we've surpassed so many of those things that I never would've thought were possible. And so when that happens over the course of 15 years, you can't help but say, "Well, yeah, I know that's crazy.
I know that's insane, but like, I don't know. I've seen other crazy, insane things happen, so, like, why not?" And I've seen them happen to, you know, myself, but I've also seen that happen to other people who, you know, society would label kind of, quote-unquote, "normal." And you're like, y- you know, you, you can do this.
And so I think a lot of what I try and do for Zapier is to like just tell those stories and to show people like, "Hey, you actually are capable of more than you, you think." And you kind of have to beat a, a little bit of the pessimism out of people. Um, you know, there's the-- Like you go read any management book, and I think this is what you're kinda t- hinting at from the Simon Sinek, is like everyone wants the like ambitious but achievable.
Like that's the, that's how we do goal setting. Ambitious but achievable. And it has to be achievable because heaven forbid we do something that we might not achieve.
[00:04:37] Jeremy Utley: Right.
[00:04:37] Wade Foster: Right. Uh, and so much of like, you know, what I am constantly doing is just trying to say, "Hey, we actually can achieve more than you can imagine.
Your imagination is too small. You need to be able to imagine something bigger."
[00:04:49] Jeremy Utley: That reminds me, I-- So you, you guys both know that I worked on a book this year, which was not on my roadmap, wasn't on my plan. I w- I was actually not planning on doing a book deal. And what happened, just, I'm just thinking about wait, here's the question: Is your strategy achievable without AI?
It's just kind of an interesting question for an organization, for an individual. If your strategy is achievable without AI, it's too small. And I learned that in an email that I forgot I sent to a CEO. A CEO had emailed me about some strategic document. This is for 2026. This is last fall. That weekend, I had received a book offer, and I was planning on turning it down because I'm just like, "I'm swamped.
I just don't have time." And I get this email. The CEO had re-replied to me kind of strongly, uh, a little bit defensively. And I, I thought, "What did I say to this person that made them reply to me this way?" And I scrolled in my own email, and what I had said the week before was, "If your strategy is achievable without AI, it's too small."
I was being critical of the strategy. And I read my own words and I was like, "Dang it." And I literally wrote my agent and I said, "We should take the book deal." Not because I thought it was possible, because the only way I would even attempt it is now that we live in an AI world. But it's an interesting question to kinda say what goal would I set if I actually believed tremendous augmentation was possible.
[00:06:13] Wade Foster: Mm-hmm.
[00:06:14] Henrik Werdelin: Can I interject one thing there? Because wait, one of the reason why at least I was very excited about, I think Jeremy too, to get you back on the pod was you've written about refounding. I have latched onto that and written probably more on refounding after. I think you probably put me onto it.
It's a perfect
[00:06:31] Jeremy Utley: segue. Perfect segue, yes.
[00:06:33] Henrik Werdelin: And I think what's interesting is if I'm just taking on your writing, it went from you calling a code red because AI came onto kinda like the scene and you were like, "This could be a real disaster moment for Sabia." And so it almost, a- as I recall that it came out of this, "Holy shit, we have to do something."
I think as I'm seeing your writing continue, it's kinda like almost moved into this more, "Holy shit, we can actually do a lot now with AI." And so it's moved into this almost like rediscovery moment of like, "Hey, this can be really exciting again. We've done it for 15 years. It's been a long slog, and we've done a lot of cool stuff, uh, probably a lot of it that we kinda sometimes talk about and over celebrate, but now we can actually do it all again because we got this new superpower."
Am I reading that correct or, or am I just projecting what I want to into here?
[00:07:31] Wade Foster: No, I, I, I'm glad that's your takeaway because that, that is very much how I feel. Y- you know, I, I have a CEO buddy made this comment, and I've repeated it over and over again now, where folks ask, you know, just like, "How, how are you doing?
How are you doing?" And my, my CEO buddy, he remarked, "You know, I, I have 10 out of 10 anxiety, but 10 out of 10 fun right now." And it really does feel like everything is just magnified. And on a day-to-day basis, like all I'm trying to do is just lean into the fun part of what we're doing because, to your point, it is, it's this like rediscovery period.
Uh, you know, I... So I live out here in Missouri, and it, it's, it's interesting dichotomy because when I go spend time with the sort of like tech ecosystem or if I'm out in San Francisco or whatnot, you sort of have this view of the world that is, you know, hyper progressive around AI. And then when I'm out here with my friends and my family and things like that, you realize how little has actually changed.
You know, what, what do they use AI for? And it's like, well, you know, they kind of experience maybe Gemini in search, and it's like, well- Mm-hmm ... or maybe they've done like use ChatGPT to do like meal planning or to come up with vacation ideas, and that's, that's really the extent of it. Like their jobs aren't any different, their day-to-day lives aren't any different.
It's just kind of the same.
[00:08:50] Jeremy Utley: Is that wild to you, by the way, just as an aside? I, like the fact that- I mean, it is kind of wild, right? ... most people's lives aren't really different.
[00:08:57] Wade Foster: Yeah. It's kinda wild, right? But, but to me, that's also where I find it so much easier to lean into the optimism side of this, where I'm just like, this stuff is, is incredible already, and yet the way most people experience it is still kinda what, like, you know, the three of us probably felt, you know, three years ago when ChatGPT came out.
Like there's still th- that's, they're still just barely scratching the surface of that. And if that's the case, then entrepreneurs everywhere should be doing somersaults because like the opportunity is still e- enormous. Uh, like we barely tapped into it. I, you know, I wasn't in the workforce in this, but the way I hear people describe like '98, '99, like, and, and not in terms of like the economics of that period, but almost like the euphoria of that period.
It-
[00:09:47] Jeremy Utley: Yeah. Yeah ...
[00:09:47] Wade Foster: it, it seems like it must be similar to that.
[00:09:50] Henrik Werdelin: But that euphoria doesn't seem to be that many places, right? So maybe could you talk-
[00:09:55] Jeremy Utley: You're saying that that was, that was, that was, uh, focused in the Bay Area even in '98, '99, Henrik, is that what you're getting at?
[00:10:01] Henrik Werdelin: I was just saying like, yeah, and I think people were more excited.
It was, I think, a more excitable time. Hmm. You know, trajectory of a, of many things- Society, world ... were going. Mm-hmm. We, it was just like the general sense, you know. I realize for a lot of people it sucked, but like the general kinda like trajectory f- felt for a lot of people was, uh, in the internet space was, was up to the right.
But I think now when, I would imagine that when you walked into Sapient, cold, cold red, there was skepticism, stuff like that. Have you managed to kind of have, make that feeling of excitement contagious in your organization? Or do you feel you're still on that journey?
[00:10:41] Wade Foster: I, you know, I, I think- The trajectory is up and to the right, but it is a daily endeavor.
And I think the, the reason it is such a daily endeavor is because of what you called out, Henrik, which is sort of you look around, like, general society vibe, and there is kind of, you know, I'd say more of a pessimistic tenor, uh, especially around AI.
[00:11:02] Jeremy Utley: Does that affect people's mindsets when they show up at work?
Oh, 100%. Do they kind- Yeah ... they kind of bring that skepticism, hesitation? And, and so what you're saying is it's a daily endeavor to kind of combat the, whether it's inertia or resistance of the outside world in, call it the Zapier environment?
[00:11:19] Wade Foster: Yeah, it's like you have to remind people of, like, you have to bring perspective.
You have to sort of bring these tools that does a better storytelling, otherwise all you hear is like, you know, we hate data centers or like, you know, negative, uh, media around big tech or, like, the, the two big AI companies or it's just, you know, a stream of negativity, negativity. I saw like yesterday the, the trailer dropped of, uh- The, the Sam Altman movie that's coming out, and you just sort of watch it, and you just sort of get a sense of like, this has an agenda to tell.
Uh, and-
[00:11:49] Jeremy Utley: What do you find is useful to counter that narrative, like in practical, everyday terms? And this, I don't know if it's related necessarily to refounding, which I feel like is, like, it- worthy of its own segment here. But how do you, on a daily basis, counter the prevailing narrative in the context of the, of the workforce, which is your sphere of influence?
[00:12:08] Wade Foster: Yeah. Well, I think you kind of just bring statistics and storytelling to, to the, the forefront. You know, the statistics give people, like, a sense of scope, but the stories are what people resonate with. And so you, you show, like, what are people actually experiencing with AI? You know, you look across, these tools are wildly popular.
There's billions of users on ChatGPT or Gemini or Claude or whatnot. These are wildly popular products. And then you remind them of like, what are you experiencing with these tools? Like, what is the most positive thing you've experienced? What's the most negative thing you've experienced with these tools?
And by and large, most people are like, "Yeah, I'm... This is actually pretty cool. Uh, this is pretty awesome stuff."
[00:12:46] Henrik Werdelin: And I think, you know, probably another thing which I, I, you know, knowing you a little bit, you seem to have the empathy to do, is also just acknowledge that there's a lot of unknown and kind of like- Totally
I understand why a lot of people get worried. I'm sure when Marco Polo was standing on the ship and going, "Oh, we're sailing that way," and like- ... if you just looked into the abyss, that, you know, like a lot of people were like, "That seems like a pretty bad idea." But I do think it is then required, to your point, to have the ability to tell a story, and I think to use energy to say, "Hey, I think it's exciting, and people who want to kind of tread along on this path, I'd be very excited to have them along."
But I also think, you know, for many founders, at least what I hear is like, and for people who do not, people who want to be very skeptical, and not that we can't have critical voices, not that we can't be worried about all the negative at- attributes of AI. I think that is very, very justified. But you also one day have to kind of say, "Well, I have a company to steward, and I want this to be successful.
I want it to be successful for our customers, but I also want it for, to be able to pay everybody's salary and make them- Yeah ... you know, money." And that requires that we basically get excited and go, go east.
[00:13:54] Wade Foster: Totally.
[00:13:55] Henrik Werdelin: Or west.
[00:13:56] Wade Foster: Let's use like a practical example of like what folks, we, we see folks experiencing.
So let's take engineers. You know, if you kind of survey the crop of like the engineering population, you won't have to go far to find folks who are like, "Is my job obsolete now?" That sort of overhang exists. And- You know, I think you kind of gotta be pretty honest with them about, like, how you sort of see the, the future existing.
On one hand, like, I think engineering job is, like, very, very safe. The, uh, number of engineering job openings is, is growing. Very few people are actually getting laid off because of this. And so there's where you have your statistics. You could tell the Jensen, you know, radiologist one that he's out there talking about all the time, like, you know, we convince people that radiology was going away, and it turns out actually we need more radiologists than ever, and now we have a radiology shortage.
Like, you can sort of tell that. But I do think you have to be honest that the job also is changing.
[00:14:54] Jeremy Utley: Mm-hmm.
[00:14:54] Wade Foster: And so-
[00:14:55] Jeremy Utley: Not,
[00:14:55] Wade Foster: not going
[00:14:56] Jeremy Utley: away- Yes, we need
[00:14:56] Wade Foster: more
[00:14:56] Jeremy Utley: engineers ... but yes, changing. Yeah.
[00:14:57] Wade Foster: Yes. It's like we need more engineers than ever, but if the thing you fell in love with about engineering was, like, this artisanal handwritten code portion of the job, and that's the thing that you yearn for I gotta be honest with you, like I do think that's a bit of a relic.
And, you know, if the new job is about trying to understand customers' problems and translating that into systems that go build and architect software on your behalf, and you don't like that version of the job, then okay, you have a personal choice to make. And so what's the choice you wanna do?
[00:15:34] Henrik Werdelin: On the refounding stuff, and I have a million questions for you.
Um, when you tell that story to the engineer, one of the issues I would imagine is that the job and the job specs and the organizational design that they sit within might be the limiting factor to unlock the energy that AI can kinda like provide, right? So I normally use this example of saying you might have a designer that can write a copy about, let's say, a dog toy.
That doesn't necessarily mean that you can redeploy somebody from the marketing team. You still need somebody from the marketing team. So now you just have like 15% of unlocked energy, but the energy doesn't have anywhere to go. The refounding that I've been working on, which I'm sure very different than you, uh, is that you almost need to kind of start with an empty piece of paper and then redeploy people into organizational designs that are built on an AI stack.
'Cause otherwise if you just try to retrofit it back into it, it doesn't work. I'm super curious to hear your perspective on that and how you're doing it at Savio.
[00:16:39] Wade Foster: Yeah. I think that that's pretty true. You know, like an example I would give is we have an autonomous support email agent experience experiment that's running right now. And so the way that that project has worked is, you know, we have a bunch of staff that has been hired over the years that provide email support, like many companies. And, you know, these folks are very skilled at, you know, taking a customer query, helping troubleshoot, replying with like a technically sound answer in a very polite and encouraging way.
They're very good at that. And, you know, as AI came onto the scene, you know, the question mark was like, well, how can that sort of assist with this job? And a lot of the first experiments were how do you augment that person doing support? And so, you know, it's okay, AI can do faster search, it can do faster troubleshooting, but yet it's still very much like AI is assisting the human to do the work.
And about a year ago, the question we started to ask was like, "Well, hey, can we turn it on its head and actually start to delegate the actual support job to the agent? And then the human's job is like architecting the system that does it." And to stand that up, it was kind of off to the side. It, it by necessity had to start there.
And so, you know, there was a small group of people, some AI engineers, some workflow designers, some smart support folks who sort of huddled up as a unit and said, "We're gonna start to work our way through this problem." And- Yeah, I think maybe, maybe counter to what, like, you would read on X or whatnot, like this is actually quite hard.
You know, delegating all your support is not something that you kind of just go buy a vendor and be like, "Great, it's solved." Like, our support is now 100% done by an agent. It took a fair amount of, like, instrumentation to sort of map out what is every step that someone takes to actually give a great support experience, and then how do we stand up workflows that address those?
Some of those are old-fashioned deterministic workflows, some of those are agents. And there's humans in the loop at, at many checkpoints along the way evaluating the output and giving, like, thumbs up, that looks good, or rejecting the output and saying, "Here's a reason why it's rejected," and that gets fed back into the, to the loop.
And so over the course of, I don't know, three or four months, that team has kind of just been working on tuning the system, tuning the system, tuning the system. And it wasn't really until like maybe two weeks ago where we actually started to get results that were high enough that gave us the confidence to start to run it fully autonomously without a human in the loop.
And even still, it's not perfect. The first attempt was like, I would say good. It's not as good as a human, but, but good and sort of in spitting distance where we're like, "Okay," like we're close, but we still have more work to be done there. And I think once we kind of get to the finish line, it still probably won't be able to accommodate every single support experience.
It will really be targeted towards a slice of them. You'll say, "Oh, okay. Ticket comes in. Okay, route this to the agent because we know the agent can handle it, but this needs to go to a human." Or, and you'll kind of just have this rinse, wash, and repeat moment where you're just steadily working bigger and bigger loops, and where the human's job is, is kind of architecting a lot of the system.
And that's in a domain like support that is one of, quote-unquote, "the easiest" for AI to tackle. It's still quite hard for it to do. And so I think if you start to go look at other disciplines, you start to realize like, oh, what used to be impossible, like there's no way you could have sort of automated this in the past.
Well, the good news is it's not impossible anymore. But the challenge is it's like still quite a bit of effort. It's still quite a bit of effort.
[00:20:14] Henrik Werdelin: Have you found places where the workflow now completely breaks? I'll give you an example. Um, let's say you apply for interns, you get 1,000 application because they use AI- Ah
to write them. Then- Yeah ... you come up with an AI system that reads all the applicants, and suddenly you're like, "This was the human-to-human workflow that I now make completely agentic to agentic," and so basically doesn't work, right? So then you have to go like, "All right, I have to come completely rethink how I'm actually doing this exercise."
Is that something you've seen?
[00:20:45] Wade Foster: Definitely. This is where I think these systems being dynamic is where I do think humans still have quite a bit of value is 'cause th- they are, they are changing quite a bit, and humans still have the ability to, like- Kind of, I, I don't know, there's just like something about our ability to see novel insights and to connect things that are somewhat unrelated that, you know, an agent...
I, I don't know, it just, it, it sort of It just falls down in some of those tacks
[00:21:15] Henrik Werdelin: At Bark we used to have this, um, we used to have this rule I called ban the bone, which is basically- Mm-hmm ... if a designer would do a design and it would have a paw print or a bone, I would basically have a little bit of a fit.
And the point wasn't that I care about paw prints or bone that much, it was that it was just the lazy. It's the most predictable design iconography you can do when you think about dogs, right? And so it just suggested to me that the designer haven't thought about it. I think AI does a lot of paw prints and a lot of bones.
Yeah. Like, it's just like here's the most obvious thing that you can come with. Yeah.
[00:21:45] Jeremy Utley: Is there, Wade, for one of the challenges that I can envision is actually an identity challenge, because humans naturally define themselves in terms of what they do, not necessarily even the outcome they achieve, right, which would be easier, I think, for redefinition purposes or, or recasting.
When you think about refounding, how does that work on an individual level where folks' identities have gotten tied up in their ability to do this handcrafted, artisanal, you know, piece of engineering? Do you just need new people who have a new identity or have, have you observed any individuals and their capacity for refounding themselves?
[00:22:26] Wade Foster: Yeah. Well, I mean, I do think this is where I somewhat am envious of, you know, folks like in, in, in college right now. Like I think the, the narrative is like, oh, no one's hiring entry-level workers, et cetera, but I think they actually have this edge, which is it's a little bit of a blank slate. And so you can come in and just learn from first principles where we're at.
Uh, now I probably, like I have seen firsthand inside of Zapier and, and in our customer base that that is not necessarily like a, an, an age thing. I've seen people who are... Shoot, I even think about my own grandfather who was like 95, and he was like, "We gotta go figure out, like, this internet thing," and it's just like- Mm-hmm
you know, so, so a lot of it is just a mindset, um, thing than an age thing. But I do think, you know, it is helpful to be around a group of people who can sort of help you do some of this way finding. Like I remember we had a, an external speaker come talk to a bunch of the Zapier leadership folks, and he made this point that in the past, how you got ahead in an organization was often because of your experiences and what you knew.
You, you were the best at this thing and you knew this thing inside and out, and because you knew that thing, you know, now you're gonna become a director or a VP and you're gonna get a raise and you'll get a bonus and all that sort of stuff. Now, when you have these AI models that are trained on all of human knowledge, what you know all of a sudden is like not really all that valuable.
It's kinda commoditized, like intelligence is commoditized. So, okay, if, if what you know doesn't matter anymore, then what actually does? And it turns out You know, your ability to be curious, your ability to adapt, your ability to learn, like those are the things that become exceptionally valuable. You can plop down in a new situation and sort of figure out, uh, you know, how to be successful.
And so if you're around a group of people who's kind of helping you shape and make sense of all these things, all of a sudden you can go, "Oh, okay, I see what I gotta do to be successful." And I, I've seen enough people do that, that I'm like, yeah, I think most of us will be successful. But if you're around folks who are just like, wah, wah, like, you know, kind of Eeyore from Winnie the Pooh, it's
[00:24:33] Jeremy Utley: just like- I think it's the, it's rare it's hard to escape that. I think most, most people, this is maybe a cynical view of humanity, most people are lazy. Most people aren't high ambition, right? I mean, we've seen it. Jeremy, maybe a
[00:24:44] Henrik Werdelin: question to you, maybe a question to you 'cause you see students all the time, right? And I, I very much kind of sympathize with or, or, or see what, uh, Wade was talking about. And I think, um, Bryce was talking about going from a knowledge worker, being someone who knows stuff, to an intelligence worker. But I think it- you're even talking about, and I think entrepreneurial people think about it- Yeah ... it's the doing worker, the one that actually shows something in output, right?
Like
[00:25:07] Jeremy Utley: a- I think there's some selection bias there, yeah. I think, I think really it's the kind of person who maybe, who, who is drawn to Zapier probably is, is more high agentic, high adaptability. I don't know. I'm just, I'm, I'm guessing. But what's your question?
[00:25:19] Henrik Werdelin: But tell me, Jeremy, when you, when you look at students, I'm sure you see the ones that come in and you go like, "They're gonna make something."
Like-
[00:25:26] Jeremy Utley: Yeah.
[00:25:27] Henrik Werdelin: Oh, yeah ... not they're gonna get like a high, you know, a, a high job in a, in a big company, but, like they're actually gonna ... What is it, what are the characteristics of that person?
[00:25:36] Jeremy Utley: Um, I, for me, what I'm always looking for, I think you c- uh, you call it resourcefulness, Henrik, but is basically is the instinct to act in the face of the unknown rather than plan or research.
That's kind of the single trait that we're l- in the, in the accelerator, for example, which I was involved in, not currently, but I was involved in for a bunch of years. We would always be assessing, is this person willing to act in order to create information, create new data, et cetera, or do they default to passivity?
And that is, uh, you know, we actually designed kind of a, uh, selection mechanism to recruit for that. We had probably w- you know, top 1% of outcomes across the university. I humbly don't think it had anything to do with our curriculum. I think it had everything to do with our selection criteria, and our primary selection criteria was we were try- we were designing circumstances where someone would demonstrate their willingness to act in the face of the unknown or not.
And that is an exceptional characteristic. Most people d- and especially kind of type A, straight A, they wanna make sure they know the rules. They wanna make sure they know what are the steps, right? And that is, it's a, the bias towards action is a, I don't wanna say it's a mark... It, it's not even maybe a characteristic of intelligence.
Maybe it's a characteristic of personality. I don't know.
[00:26:59] Wade Foster: Do you think that that is... W- well, I'm curious, do you think that that is innate to the individual, or do you think that that is shaped by the environment or- Yes ... uh, the work that they are doing? Both. Both. Yeah. I mean, that makes sense to me. Yeah.
Yeah. I, I, that's kinda what I think.
[00:27:14] Henrik Werdelin: But I do think many organizations probably, to Darra's point, and maybe back to the refounding, if you're gonna refound your company, if you're gonna make it... I have companies that I built, uh, before AI and after AI, and they're just completely different companies. Like, it- Mm-hmm
and people are gonna know what, what really, it just, it's just everything. You know, like-
[00:27:31] Wade Foster: Mm-hmm ...
[00:27:31] Henrik Werdelin: one that's built after we deploy code every six hours. That is just the rhythm of the company, right? The other one has a roadmap, right? Like, and, and so I do think that it is very, very complicated to change the environment.
If the mood of the room have been, "This is how we do stuff," and the people who have a senior position now got a senior position because they knew how we did stuff, then, I mean, like, it's just very complicated. And so, uh, uh, that's why I think the refounding term is such a powerful term because it is a little bit about the forest fire.
[00:28:05] Jeremy Utley: It's making a statement about the context. And, and- Yeah, and- ... to your, to your question, Wade, Dan Pink was on the show, and we were asking him about agency, and he actually said, most of the time we talk about agency as if it's a personal trait. But he said agency is a function of environment And a better question than is how do I become more agentic, which is about like individual whatever behavior, is actually as a leader, parent, teacher, et cetera, how do I create an environment where agency is the norm?
And I thought it was a really beautiful kind of, uh, uh, reframing, right? That agency is about context. And then as stri- A- and then as shapers of context, whether we're teachers, leaders, et cetera, we actually have an enormous impact on someone's agentic capacity. Which brings me back to refounding, right?
It's like this is maybe a chance-
[00:28:54] Henrik Werdelin: And I think that goes back full circle to like a lot of people we interviewed on this podcast that are successful, they did something that was not in their job specs, right? You know, we talked to the guy who does hardware for Meta and like he redesigned the Google homepage.
Or I seem to remember. You probably, Jeremy, remember better. The, like, all the people... And, and like you probably have that the people in your organization that you probably give the most attention to, honestly, are the people who go like, "Hey, I also thought about this, this, and this, and by the way, I coded it up and here's an example of it," right?
And so I do think the future belongs to a workforce of people that has resourcefulness and understands these new tools. And I think maybe is, to your point about college kids, has not been trained in that you have to stay in your lane.
[00:29:39] Jeremy Utley: Because now with AI- But that's, but that's what, that's how college is training d- dude.
That's part of the problem. Like I, I took two, two young guys from my church to breakfast the other day. This is a true story. They're both computer science studying in the UC system. I'll leave it at that. And they were both asking me, "Hey, the job market's tough. What do we do?" And, uh, one of them wanted to apply for a position at Roblox.
And I said, "Well, tell me about Ro-" And he's, and he's like a passionate gamer. And he started talking about what's wrong with it. And I said, "Why don't you build an app?" And he said, "That's what I'm doing, but there's like thousands of applications." I said, "No, I didn't say write an app. I said build an app." And he looked at me like I was speaking Swahili, honestly.
He looked at me and he's like, he said, "Uncle, no one has ever told me to give my own assignment." Like, and I realized there we are training ag- The, the current e- uh, you know, system of education is teachers give assignments. And the reason students want jobs is because they need someone like a teacher to tell them what to do.
Because all of their education they've been conditioned to look to the teacher for what do I do. And then they w- You know why they want a job? 'Cause they need somebody to give them their next assignment, right? And they never learn to commission themselves or to assign themselves.
[00:30:50] Henrik Werdelin: Wait, do you think it's the same in organizations?
And if yes, how can you tell people that that is not the case?
[00:30:58] Wade Foster: It definitely, I mean, humans, I guess where humans and AI are a little bit alike is that we're, we're goal-seeking or we're status-seeking, I guess, in the ca- case of humans. And so there definitely is this desire to do well, I think. I do think most humans want to do well and want to succeed, and I think what a lot of folks are trying to figure out is like, what are the rules of the game that sort of allow me to be successful?
And that's where I think, that's where the, that's where the playing field gets interesting, I guess. Um, because there are, you know, there's sort of like a set of general advice, and the masses probably continue to get coached on this general advice, and as a result, their output is general, which is also average.
Oh, you should apply to a job, right? Better advice is like what Jeremy said, was like, "Hey, you need to stand out somehow." And it sounds like you actually do have some, some keen insight. And if you try and go through, like, the general advice, like you're gonna look average. But if you have a different insight, go play that game.
And now what's interesting is now you've given somebody the ability to think about it differently. And I, I mean, it'd be interesting to see what, you know, that student does with that advice. 'Cause if, if that's all it took, like that's probably a pretty high agency person, and it's sort of an indictment on, you know, probably a lot of like university like career coaching centers where it's like, "Hey, you're just out of whack with how the professional world hires people."
You're coaching them on a dated mechanism. And so, uh, I don't know. It's, it sort of feels like you've got, you know, the systems problem on one hand, which is like, "Hey, how do we actually just do a better job of like coaching folks up to deliver on the things that we care about?" And then you have the individual advice, which is like, "Hey, you know, how do you just
Uh, I mean, I kind of just come back to my YC days where it's like you kind of just want to figure out how to make something people want, and that sort of works in, like, all aspects of life, where it's like you as a human, if you kind of just know what the, like, one nugget is, and then you can figure out how to apply it across all areas of your life, that sort of feels like where you almost get this, like, this, this cheat code where you're just like- Mm-hmm
oh, I can walk into any situation even though I'm unfamiliar with it, but I kind of have this, this apparatus in my head that, like, allows me to, like, quickly figure out how this works and go like, "Ah, now I know what I need to do."
[00:33:28] Henrik Werdelin: But I think that's such a beautiful point, which is, yes, you have to find out.
You just make something that people want. But I think there's almost this step before where the, the line i- ex or after is like, and you can just do it, right? You know? I-- we, we did this, uh, at Autos. We have these people. We help people build a company from scratch. And we did this summer camp, and we had hundreds and hundreds of people go through this camp, and one of the things that kind of dawned on me, a lot of these people kind of came in and then you said, "And then by the end of these 12 week, you're gonna have a startup."
And they're really like, "Really?" "Yeah." It's like, you know, "Can, can I just do it?" It's like, "Yeah, you, you, you can just do that," right? You know, like, nobody's gonna prevent you from buying a domain or coming up with a... seeing a problem that a customer have and then make a little authentic solution for it like that.
We, we will get this done in 12 week, right? So, and the excitement and I, I think the life-changing this has been for a lot of people, just kind of realizing that there's no magic to it. You just do it, and then it's there, uh, is quite profound. And so maybe that's part of this refounding thing is honestly also what we have to teach everybody.
Not we have to teach as like we know the answers, but like, which is one of the things that a lot of people in the workforce, I think, will really benefit from understanding.
[00:34:43] Jeremy Utley: Okay. Refounding, though, can we define this? 'Cause we've been saying this word a lot. And by... And audience notes, the reason Henrik and I invited Wade back to the show is because he made a post about refounding Zapier.
We've danced around it a thousand different ways. Wade, what does it mean to refound Zapier, and how is it going?
[00:35:06] Wade Foster: Yeah, I, I, you know, the thing that I was trying to convey to the team is that, you know, the things that made us successful in the last 15 years may not be the things that make us successful going.
And so this is your open invitation to rethink anything inside the company. Rethink our products, rethink our way of working, rethink down to, like, even, like, this, like, micro small stuff. I'll give you a dumb one. Like, I had a signature request come across that wh- while we were sitting waiting, and I was like, "Why am I having to sign this?
This is, like, dumb red tape." And so I just fired it off to the person in charge. I'm like, "Why are you asking people to do this? Can we just get rid of this process?" And so it's just, like, inviting people in to say, "Hey, let's go rethink all this stuff."
[00:35:47] Jeremy Utley: It's bestowing agency- And- ... in a way. It's, it's, this is a public pronouncement.
You have full- Mm-hmm ... agency to question everything. Is that effectively what you're saying?
[00:35:55] Wade Foster: Basically, and, a- and then, like, allowing people to sit with the discomfort of what is about to come because we are going to reinvent a lot of what the company looks like. And that can be a jarring experience if you sort of show up...
You're, you're sort of used to showing up to work and sort of going like, "Okay, I know what my day is gonna look like and da, da, da." There's sort of, like, a personality type that really likes that routine. And, you know, this is a little bit of just, like, saying, "Hey, here's what the new routine is gonna look like.
And you are gonna show up to work, and some of these things are gonna change, and things are gonna change pretty quick." And, like, this is actually part of what it looks like to be successful and to reorienting people, um, to, to that new world. And so that's kind of the, the, the mindset shift that I was trying to instill in the company.
[00:36:40] Jeremy Utley: Is that supported by mechanisms? For example, like a PR FAQ, Amazon's kind of famous mechanism. But I can imagine if you say, "Hey, question everything," people go, "Where and how- ... and to whom?" Like, what kind of structural support do you have to give people who have now been told they can question everything?
[00:36:57] Wade Foster: Sure. You know, I think th- this is where, you know, I do provide, like, a little bit of guardrails. And so it's like, hey, here's kinda like the updated goals for Zapier, updated strategy for Zapier, updated operating rhythms, and, like, here are some things that I think are, like, really important for us to go shift.
And so it kinda, it kinds of shine a light on the areas that I think are really ripe for it, and that helps... I don't know. It's like, it's like b- bowling. It's like you kinda put the bumpers up, and so it's like, all right, like- It's
[00:37:25] Jeremy Utley: kinda like the hackathon. So, so Wade, we had, we had a guy, uh, Nick, who was on a couple weeks ago.
He's run, like, 3,000 hackathons for NASA. Yeah. Yeah, your face was exactly right. That's a lot of hackathons. And one of the things he says is you kind of want some vectors. They, they don't have to be overly specific, but for example, we're trying to positively impact kids' health. You know, we're trying... Or kids' nutrition or mental health or after school safety, right?
But like broad kind of vectors and then you go, go that way. Is, is that what you mean basically by the bowling lane, so to
[00:37:57] Wade Foster: speak? Yeah, basically. Yeah. I mean, it's the, uh, there's like a term for this, but like when you prompt somebody a question, when you say, "Hey, I want you to think of all the white things that you can think of-" Constraints.
Yeah. "... uh, you usually get the same number as if you say, think of all the white things in your fridge." Yes. And it's like, well, surely the former has like a much larger number, but for whatever reason, humans tend to answer with the same amount for both. And so like, you just kind of give this like focusing of it, and it doesn't like prevent them from sort of working outside the fridge, so to speak.
But it, it, it does provide like a little bit of direction.
[00:38:31] Jeremy Utley: What were the vectors that you are giving people?
[00:38:36] Wade Foster: Yeah, I think for us it's like we are going to be great at automation. That is the thing that Zapier is, like, really known for, and we will continue to be known for. But how we go do that, we fully believe that it's not gonna be a human that's building automations anymore.
You're not gonna go click a trigger and click an action. It's like you're gonna go talk to an agent, and the agent is gonna be the one that go, sort of assembles this. But then, you know, how does that get run? How does this thing... So, you know, a lot of it's like providing some guidance on like, "Hey, here's sort of how we view, like, the, the shape of automation of the future, the shape of things to come, and, like, where the important shifts are happening in society," and then just letting smart people kind of pick up on those themes and go like, "Ah, okay, I, I kinda get where we wanna go.
Let's go innovate within those mechanisms."
[00:39:18] Henrik Werdelin: Here's my last question for you. You are now a few years into your refounding kind of journey. For the people who listen to this, who sits there in the company and go, "Hey, this is a refounding moment for us, too," what's on your cheat sheet for how they should think about it?
[00:39:35] Wade Foster: Well, yeah, I think the most important thing, I, I, I mentioned it once before, the YC mantra, make something people want. And, you know, if you're refounding your company, big part of it you need to just be really obsessed about, like, what are your customers, who are your customers, and what are the things that they want?
And we sit at this moment in time where AI has just disrupted so many of the traditional ways of building software and selling software and finding soft- and all that sort of stuff, that that has created an enormous amount of opportunity. And so my guess is for people who have companies that are older than ChatGPT, you should be saying like, "Hmm, how can we refound, or how can we take advantage of these new paradigms, these new ways that people are finding work, they're building things, they're making stuff happen inside their company, and, like, where can we seize an opportunity there?"
[00:40:32] Jeremy Utley: Is that distinct? One of the questions that came up early on, like I remember JJ Zhuang, who's the, uh, chief architect at Instacart, for example. He's one of our first ever episodes. And maybe it was David Okneyev, the founder of Typeform. One of the two, they said, "You have to go through the exercise of what would we be if we could be founded post-AI?"
Is it the same question or is it slightly... It, it feels like maybe it's slightly different. I don't know if you have a thought on that.
[00:41:00] Wade Foster: Oh, I don't know is the honest answer. Like, I think that's a useful question. There are certain prompts that you can ask yourself that just kinda get you thinking about things a little bit differently and say, "Hmm."
Uh, you may not wanna start totally from a blank slate because you have assets that are, like, really valuable, and you wanna take advantage of those. And so if you were truly starting with a blank slate, you wouldn't get to take advantage of those. But the prompt itself can be revealing because you'd go, "Hmm, what would the, like, YC startup do that was kinda attacking this space, and how might they go about it?"
And it's like, oh, but what if they also had access to this, this, or this? What would they do then? Um- Hmm ... and so, y- you know, I do think those, like, lines of questionings do kinda help get you outside of, like, maybe a mental container that you've sort of find yourself in.
[00:41:45] Jeremy Utley: That's the famous, uh, thought exercise that Andy Grove went through at Intel.
I don't know if you know this.
[00:41:50] Henrik Werdelin: Mm.
[00:41:50] Jeremy Utley: But famously, he and, I don't know if it was Noyce or o- one of the, you know, founders there, they were in the boardroom, and they had this big decision about w- which direction to take the business in. And they're, I think as the story goes, they're overlooking Great America, which is kinda like this, you know, a park, like a theme park.
And Andy said, "If the board fired us both and brought in a new guy, what would the new guy do?" Hmm. And he said instantaneously it was... And there, that, that's actually, I learned about that from a book called Psychological Distancing, right? Uh-huh. You can be someone else, someone else, or somewhere else, and a lot of times that will jar your perspective.
But- Yep ... by asking themselves the question, what would the new guy do, it was abundantly clear what the answer was.
[00:42:32] Wade Foster: I like that question. I think of that often. Uh, Frank Slootman's a, like, one of the most world-class CEOs, and he's, he's a hired CEO. He always says, "I'm not a founder." And I, I ha- I was fortunate enough to be at a dinner with him, and I remember, you know, thinking out, at that dinner, I'm like, "I wanna do all the things that Frank would do so that we never have to hire Frank."
[00:42:52] Jeremy Utley: It's brilliant
[00:42:53] Henrik Werdelin: I think that is an amazing point to maybe stop the episode. Mm-hmm. Hey, Wade. I mean, like, I get so inspired every time we talk together. It, it's really, really impressive what you're doing, not only with your company, but also how you're doing it. So thank you so much for, for giving a lot of us, uh, a bunch of inspiration to, to bring to our organizations every day.
[00:43:12] Jeremy Utley: Well, thanks for having me on. I ask myself what would- You guys have a great pod. I ask myself, what would Wade do? That's what I
[00:43:16] Henrik Werdelin: ask. Yeah, what
[00:43:17] Jeremy Utley: would Wade do? You're my Frank Slootman. You're, you're my Frank.
[00:43:22] Wade Foster: Oh, too kind, guys. That's awesome. Well, thanks for having me. Y'all are a fun to... fun podcast to come on.
[00:43:26] Jeremy Utley: This is great. I got so much stuff.
[00:43:30] Henrik Werdelin: You got so much stuff. Hey, um, go, 'cause I, uh, have a full-on man crush on Wade, so I could also go on for a while.
[00:43:38] Jeremy Utley: You wanna gush? You wanna gush for a minute there, Henry? Yeah. Yeah. You know, we were just saying between, it's fun. Uh, there, there's a certain kind of energy that comes from a fellow builder, uh, that's, you know, unmistakable.
Very cool. I thought, uh, I'm just looking at my notes here. You know, the point of... that he was making about roles. I- is the job going away? No. Is it changing? Yes. I thought it was a very simple and elegant way of expressing what's happening right now. And if you're not comfortable with change, that's gonna be a problem.
You, you don't have to be fearful of displacement or being fired, but the r- the job's gotta change. Yeah. And I think just stating that plainly is, is quite helpful.
[00:44:16] Henrik Werdelin: I think that this new type of person that is required for the future company is a fascinating kind of concept that I'm haven't seen too much writing about.
Like, what is it that this person have? Mm-hmm. And I think, you know, as you know, resourcefulness, doing more with little, or doing a lot with little. I think Melassams that you mentioned in the pod, which I thought was a very astute point, that at the incubator at Stanford, you know, people who kind of don't stop when they don't have an answer, but they kind of use making as thinking as a mantra to work themself into whatever is the next action, um, I think is, is fascinating.
And then there is this bias for action, right? Somebody who just do, and probably even is a little bit untethered and untrained, so they sometimes even do in lanes that they're not allowed to, or they previously was not allowed to.
[00:45:14] Jeremy Utley: They, they don't know what the prior art is. They don't know what the right answer is.
You know, so many times breakthroughs come from kind of thinking outside the box, thinking outside the lines, and the less familiar you are with the art of... the prior art, the more capable you are of doing something unexpected.
[00:45:28] Henrik Werdelin: And you know what, when he was talking, there is this folklore in organizations.
Like I, I was sometimes reminded at Barbux, people would say, uh, they'll show me something and go like, "Oh, I'm very sorry," like we used Green. And they go, "And I know you hate green." And you're like, "I hate green?" And you go like, "Yeah, I remember like, you know, a few years ago in a meeting you were making a whole thing about, like, how green thought."
And it was like, uh, I think that was a very specific thing for that specific day about that specific thing. This was not a general kind of perspective about the color green. I do think that these folk laws are in our organization, and so what I really enjoy him talking about was we should just question a lot of these things.
This policy that we've always done, like why did we actually set it up back then? Yeah. And in the age of AI, or just in the state where we are now as an organization, is this really needed anymore?
[00:46:17] Jeremy Utley: The other thing that I thought was interesting on the topic of refounding, which it felt like we danced around more than we talked about.
But I think the, the... Basically what I took from his, the underlying principle of refounding is making sure everyone's clear they can question everything. And to me, I wonder if that could be a permanent state, if it has to be kind of a temporary thing, meaning it's a point in time where everything is questionable.
Or if you could actually just say everything is always up for debate, up for grabs. It- Why was it that it was a refounding moment is my que- Uh, is it just, it's a good excuse?
[00:46:57] Henrik Werdelin: Yeah, and I think there's like a subtlety to it though also, though, because what you're not looking for is people just saying, "Oh, I'm just asking question.
Like, I don't know, like, you know, I don't have a solution. I'm just, like, asking the hard questions." Nobody need that person. What they need is, "Hey, we've always been doing it this way. I thought about this other way of doing it. It doesn't seem to be within the parameters of how we've done it before, but it seemed to be smarter, and it's not more expensive to do."
N- nobody ever said, "Let's not do that," right? You know? Um, and so I d- I do think that the reason why refounding is tough to talk about is that, like Montessori, a pedagogy of teaching, it's tough to say, well, what are you gonna do on a Wednesday at 2:00 PM in third grade? There's not necessarily, like, a specific formulaic kind of way of doing that.
I think it's the principles, a way of thinking about how to do stuff in a new way and applying a certain set of principles, which I think for me at least, what refounding is the umbrella of. And that is more some of those things like question, but don't just question. Come up with a solution, probably using AI and, and those kind of elements.
Does that make sense?
[00:48:07] Jeremy Utley: Yeah. Yeah, I wonder- I, I think we didn't dig into this with Wade, maybe it's, uh, episode three, um, you know, six months down the line. I have an open question in my mind generally about-- not about Zapier particularly, although it's an interesting use case, but generally speaking, how comfortable are people questioning what is?
Like defaults are a, are like... If you think about behavior design as a field and in behavioral economics, one of the primary findings is defaults are powerful. So simple example, very simple example, in countries that are very similar, I don't know, say, uh, Denmark and Sweden, are they relatively similar? I don't know.
[00:48:47] Wade Foster: Yeah.
[00:48:48] Jeremy Utley: They may have radically different rates of, say, participating in donor registries. Usually the reason for that difference is because the question is like what's the default on the form? Is it check this box to opt out or check this box to opt in? And it turns out- I
[00:49:05] Henrik Werdelin: heard that somebody's talk about it as the devil's in the default- Yeah
which I think is such a good... Yeah.
[00:49:09] Jeremy Utley: Yeah, it's, it's perfect. And in the country with a high participation rate, it's the default is you're in unless you choose to opt out. The countries with a low participation rate, the default is you're out unless you opt in. And what designers realize is with really hard decisions, people just don't change the default, right?
And so- Yeah ... it's like, so then as a system designer you go, "Well, what's the outcome we want?" Why do I mention that? The environment itself is the default, and when you say question everything, in a way you're going, "Don't take the default as given." Well, just from a behavior design perspective, that's really hard.
The default is actually probably one of the more controlling environmental factors. And so to say- Mm ... "Hey, imagine you didn't have to be in that studio. You could be anywhere." You're like, "Well, I can't 'cause I'm here," you know? And, uh- Yeah ... so all that to say, I wonder... I think it's wonderful to say question everything.
I wonder how you s- how behaviorally from a ritual perspective, norms, et cetera, how you reinforce. Like I, I bet my hunch is Wade needs to celebrate people who question things regularly for everyone to see, oh, it's really okay to question that.
[00:50:19] Henrik Werdelin: I will say that I've seen increasingly amount of people who are doing this, and while I think somebody like Wade is doing on an organizational level throughout all of Zapier Most of these initiatives that seem to be the breakthrough initiatives seem to be happening when you break out of the environment.
You know, he was mentoring a team where he put like basically a SWAT team to do that.
[00:50:47] Jeremy Utley: Mm-hmm.
[00:50:47] Henrik Werdelin: Mm-hmm. We talk about the hackathon as a vehicle, an environment vehicle for change. I think, you know, some people kind of, at least like some of the projects I've been involved in, we do set up a new instance and then we try to build it from scratch.
I do think there is, at least it is way more difficult to do it kind of like on a whole organizational level, and I think it almost have to be built up that way. I would imagine that NVIDIA didn't suddenly become just very entrepreneurial. I would imagine that they had been like that for a while
[00:51:25] Jeremy Utley: Yeah, and there have to be mechanisms to reinforce that.
So like for example, at Amazon, if you write a PR FAQ and it gets approved, you're immediately out of your previous job and into the do this thing, and they've got
[00:51:40] Henrik Werdelin: people- But isn't it crazy that that is the example that we always use?
[00:51:42] Jeremy Utley: Yeah.
[00:51:43] Henrik Werdelin: Like that there isn't 200 examples like that. And you're like, "And then at Microsoft they do that, and at Google they do that, and at Mail they do that."
Right. There just isn't a lot of people who thought about, it seemed like how do we change the environment so that this becomes a more default path?
[00:51:57] Jeremy Utley: Mm-hmm. Mm-hmm. Maybe that's a cool, um, request to the audience. If you know of stories of organizations with structures that drive- Absolutely ... exploration and experimentation, we would love to talk to them.
[00:52:10] Henrik Werdelin: You know who's do that, where I've seen it also is in the SEAL team. They do get these very difficult challenges, like you have to go in and grab this thing, and then basically come up with a way of doing it, and then they seemingly have pretty wide range of tools to be able to do that. And they do think like that, right?
Like it isn't that they go like, "And last time we did this, we did it in this way." You know? Mm-hmm. "This is how we always do this." I, I get the sense that they kind of go, "Okay, let's figure out how to get in this way." So at least like that's one ar- other area where I've seen they've created that environment.
[00:52:45] Jeremy Utley: Do we know any I may. I may be able to get somebody like
[00:52:50] Henrik Werdelin: that I may. I may. Awesome. Anything else we should add before we let people go?
[00:52:56] Jeremy Utley: You know, I mean, they'll, they'll... I'm just looking at all my notes here. Um, I love, like, having a what would Frank Slootman do? I think, like, that the insight or the remind- the, the memory I had about the Andy Grove story, I think it is a very helpful tactic.
Asking yourself, what would someone else do, oftentimes is illuminating, so that's just like, you know, a, a takeaway tool anybody could use immediately. What would Wade Foster do? What would Frank Slootman do? What would Henrik Werdelin do? Those are great questions. Just kind of break out of your rigid patterns.
And then the last thing that I would mention is s- one thing Wade said at the beginning, which is he's living in Missouri, out of the bubble, certainly out of the Silicon Valley bubble, and he said most people's lives aren't very different. And it's just, it's an interesting reminder. I mean, I would say, like, you know, I do keynotes all over the place.
Most people I talk to are on the very beginning of their journey. And so, you know, we are immersed in these worlds and these conversations with people, like, hyper-augmented everything, and most people are, uh, going, "Should I pay for ChatGPT or just keep using the free version?" You know, that's like, that is the- Yeah
vast majority of people in the world are still trying to figure out, is it even worth it to pay? Yeah. And that was a, I think it was a very humbling and, and timely reminder in this. I mean, we started the conversation by talking about should we just shut model development down because of all this stuff, and most people aren't even paying for AI.
Yeah. And our audience probably is prone to forget that.
[00:54:24] Henrik Werdelin: And with that, I think it's time to end a very, very good episode, I would say.
[00:54:32] Jeremy Utley: Dude, we've got quite a week here, I would say. I mean, we've got, like, four- We've got a good- ... like, banger interviews this week. I know. Just get ready, folks. Yeah. Get ready.
[00:54:39] Henrik Werdelin: And with that, the only thing that's left to say is to say
[00:54:43] Jeremy Utley: bye-bye.