The Top Mistakes Made by Startup CTOs with Zach Goldberg, CTO at Gruntwork

(Intro Narrator at 00:00:00) Today, we're talking to Zach Goldberg, executive coach and CTO at Gruntwork, about his book, The Startup CTO's Handbook, and the nature of AI ethics. You're listening to Joel Beasley, Modern CTO.

(Joel Beasley at 00:00:18) I'm excited. The way I found you is I was just scanning through, I think I saw your Startup Handbook on Amazon. When did you write that?

(Zach Goldberg at 00:00:28) It was published October 16th of last year.

(Joel Beasley at 00:00:31) Why did you write this book?

(Zach Goldberg at 00:00:32) Why did I write this book? I've been a startup CTO for the better part of a decade, and I think in particular, at that time, I'd been with the company for three years. I was working with a number of other really awesome leaders, and I just felt we had created and captured so many ethoses and processes and ideas for what is good engineering leadership. And that team in general had a really good distillation and ability to convert these ideas into actual practice. And so I just had all this stuff in my head, and I was like, you know, I'm not seeing this in blogs. And so I made up just a bulleted list of all the different things that I thought was interesting that I had seen over years and was fortunate to have a group of people to bounce those ideas off of. And that outline, I ended up being without work for a few weeks. I was like, you know what? Time to actually make this happen. And so that was the genesis of the book.

(Joel Beasley at 00:01:24) And then now what do you do?

(Zach Goldberg at 00:01:25) Now what do I do? I am the CTO of a company called Gruntwork. We make DevOps infrastructure platform tools, and I am an executive coach to other technical leaders.

(Joel Beasley at 00:01:36) Oh, that's cool. So you're actively coaching technical leaders?

(Zach Goldberg at 00:01:40) That's it. Yeah.

(Joel Beasley at 00:01:41) Okay. What mistakes are people making? What are you seeing right now?

(Zach Goldberg at 00:01:46) Goodness. We're all human. We all make mistakes, right? I think it's not necessarily that people, I wouldn't call them mistakes. I call them learnings. They're gaining experiences. It's moving from being a technical person to a manager, right? It is a giant transition, far larger than I think anybody realizes it is. And you're suddenly in this position where, A, you're the technical expert, but you've never been in a position of organizing people before and organizing business process and being responsible for business strategy. A lot of your career as an engineer is you're given a technical problem to solve and you become an expert at solving technical problems. But the reality is that's not why businesses hire engineers. Businesses hire engineers to use technology to solve business problems. And that separate link of why are we building this and how does it move and generate value for the business is often a product manager's job or an engineering leader's job. It's often delegated elsewhere. Now suddenly you're that person. And so there's all of these additional inputs and different types of thinking and organization and process to put in place to make sure that you're actually seeing it through all the way to that end business value. And so a lot of what I see is, and a lot of what I work on with these leaders is that gap, right? And how do we, I'm not going to teach you anything about technology. I'm not going to teach you how to write code. Presumably, the reason we're talking is you've already demonstrated that you're great at technology. But how do you hire people well? How do you set goals and set expectations with the team? Performance management, you know, all of these sorts of topics.

(Joel Beasley at 00:03:20) That's pretty cool. And so how many clients do you have currently?

(Zach Goldberg at 00:03:24) I'm working with five leaders. So CEO, director of engineering, three CTOs.

(Joel Beasley at 00:03:29) That is pretty cool. Now is that CEO, were they a software person before?

(Zach Goldberg at 00:03:34) They were indeed. And the genesis of our conversation was, we're having, you know, we want to level up our engineering team. And it just sort of turned out that we had a, we developed very quickly a good working relationship. There was stuff to work on beyond that, and so we continue to work together. One of my favorite clients, actually. He's fantastic to work with. It's very satisfying to have somebody come to you and say, this is what's going on in our business, work through it with them, and then still be working with them a couple months later and see the impact. Right? I'm not actually doing the work, of course. It's all, the credit goes to them. But that requires that somebody shows up with humility, a desire to learn, put in the effort, and really level themselves and their organization up. And to watch that happen over time is just profoundly gratifying.

(Joel Beasley at 00:04:19) Have you found that it's helpful to mentor these other people at these other companies in your own personal growth?

(Zach Goldberg at 00:04:26) This is a dirty little secret, Joel. Dirty little secret is that, yes, interacting with other leaders every day makes me better, right? It makes me a better coach. I can, you know, lessons and patterns we see with one client are helpful in providing context and perspective to others, and then inform and advance my thinking on how engineering works. And, of course, I bring that back to Gruntwork and vice versa. You know, what I see on my day job working with my team at Gruntwork also informs and provides perspective to the clients I work with.

(Joel Beasley at 00:04:54) How large is Gruntwork currently?

(Zach Goldberg at 00:04:57) Depending on how you count, let's say Gruntwork's 20 people. So we're an interesting company from an organizational chart perspective just by virtue of the product that we offer. But we're an eight-year-old bootstrapped startup is how I would describe it.

(Joel Beasley at 00:05:10) Oh, are you one of the co-founders?

(Zach Goldberg at 00:05:12) I am not. I was brought on earlier this year to lead the tech team.

(Joel Beasley at 00:05:16) Oh, cool. Very cool. And what were you doing before that?

(Zach Goldberg at 00:05:20) What was I doing before that? I was the CTO at a company called Equi, which is in the investment tech space, on a more traditional venture track.

(Joel Beasley at 00:05:28) Okay. Very cool. Yeah. Lots of similarities. I did some fintech-type software and yeah, I found that when I was helping, I did some code mentoring for a while when I was a full-time software engineer, and I found it was very helpful that I had people I looked up to that were better than me. I had people that were my peers, and then I had people below me who I was helping up. And somebody had, I read a blog post or a book or something that someone said, hey, look at it like a cross. You have the people above you, your peers, and the people below you. And then you'll feel a sense of completeness if you are managing these types of relationships in your life. Because I don't know if other people do this, but I tend to just lean towards one, right? I'm either always just hanging out with my peers or I'm only interested in what people are doing ahead of me. So I started writing those letters to the professors. I'm like, hey, I want to get involved and come speak to your college students. And I want to, you know, I've been reaching out. I started an email campaign last week to reach out to CTOs, be like, hey, just tell me what's going on in your work life and just trying to engage with the community more.

(Zach Goldberg at 00:06:30) Yeah. I think there's, I completely agree with you. There's definitely a sense of, you know, who you interact with absolutely informs that perspective. I have this problem where for a period of time, I was very focused on reading business books. And so I have a list of all the business books I've read. And one nice thing about it is a lot of the top books have very common themes, right? Like focusing on culture, supportive collaborative environments, these kinds of things. And yet still, I find that your brain can only hold so much. There's a huge recency bias to your decision making and your perspective. The most recent book I read is when I'm mentoring somebody and I'm quoting an author, I guarantee you nine out of 10 of those quotes are from the most recent book I read at any given point in time. I just finished the Netflix Reed Hastings book, so I'm going to quote that one constantly to you. And so there's the sense of that's just what it is to be human. I can't remember, and I can't recall all the valuable lessons from every job I've ever had and every book I've ever read. And so the way you compensate for that is to make sure that you're constantly getting input from this diverse group of sources that supports this type of perspective you're trying to build. And so, you know, that means for me continuing to read, right? So there is always a fresh book, working with these clients and at work. And the combination of all these things together sort of puts the good ideas and keeps them fresh, right? It's like, you know, it's like RAM. I need to constantly rewrite the data to my RAM. Otherwise, it slowly gets, it slowly fades away.

(Joel Beasley at 00:08:02) It's like muscles, like working out every day.

(Zach Goldberg at 00:08:05) Yep. The brain is a muscle. Absolutely.

(Joel Beasley at 00:08:07) I started last week working with another engineer. I haven't written an application in five or six years, but I wanted to put our thousand episodes into an LLM to be able to talk to it. And so I've been going down that path, and that should be ready next week. So I'm excited because to your point of reading books and diverse people, I have a body of work that's very diverse over 10 years that I would like to engage with, because I can't remember the conversations. I have people that meet me in public and like, man, it was so great being on your show. And I was like, oh my goodness. We know each other. Hi. Because it's just so many conversations.

(Zach Goldberg at 00:08:43) Absolutely. Yeah. How is that going? What have you, have you interacted with the LLM yet?

(Joel Beasley at 00:08:49) I have on a single episode. So we trained it on the Mastercard episode at first, and then we learned about embeddings and vector data stores and the RAG-type searching and how we're going to get stuff out with it being such a, I mean, we got 40, I think it was like 42,000 hours of content or something like that.

(Zach Goldberg at 00:09:09) Oh, goodness. That's nothing for modern GPUs, of course.

(Joel Beasley at 00:09:12) Right. Right.

(Zach Goldberg at 00:09:13) Right.

(Joel Beasley at 00:09:13) Well, we've done all the speaker identification and topicization. There's all these little things to it. And then I went around yesterday on LinkedIn, and I messaged like 20 people. I was like, hey, what would you ask this LLM? And then so I made those test questions, and we're going to get to play with it a lot more tomorrow. So today, we're actually training it on the remaining episodes, and so I'll keep you posted on how that goes.

(Zach Goldberg at 00:09:39) It's very exciting. Yeah. I've got a friend of mine who the way he learns and experiments is he likes to do it in public. And so I've got emails from him twice a week for the past several weeks. It's like Alexei's journey through building custom LLMs to help himself in his life. And so he's like, this is my problem statement. This is how I'm thinking of using an LLM. And then I get an email three days later. This is what I tried. What advice can I have? And so, you know, over the progression of a couple weeks, he's helped file his taxes and organized his contacts and created an LLM that's a CRM for his own personal life. And so I think it's great that we're, you know, the question is, what's this look like in two, three years? Right? Four years, five years? And all of these custom use cases right now, it takes somebody, perhaps with your level of expertise or Alexei's level of expertise to, A, identify the use case and plug the bits together, you know, wire it all up. But that's just going to be by default. All of these things will come in pre-canned forms in one way or another, and I'm excited to see that, you know, for the audience more broadly.

(Joel Beasley at 00:10:46) I fully agree. The thing I'm watching and I'm particularly excited about is the startups coming out right now, like Venice.ai, the ones that are privacy based. So their whole thing is, you know, we're not storing your information. We're not using it to train better models, and they have like matrices of the different vendors out there and their proprietary models and all of that. And so the, let's call it the freedom market, open source-type market, is figuring out how to create an ecosystem. And they're doing it. But it's like a year behind. So you can go interact with these tools, and they're all a year behind. They lack APIs. They lack certain things. But the fact is that there's money there and it's growing. So I'm like, yeah. I'm not going to not build until privacy is available. I'm just going to accept the cost of lack of privacy until such time as I can just port everything over. And I estimate it'll be two to three years before they really catch up and are in line with them.

(Zach Goldberg at 00:11:53) Yeah. Absolutely. I am, I'm in that vein. I'm very excited. You know, I don't know if you watched the Pixel 9 reveal the other day from Google.

(Joel Beasley at 00:12:02) Their phone failed? I saw the clip on that.

(Zach Goldberg at 00:12:04) That was pretty great. They got it eventually. But it does seem like, you know, even Google is pushing for on-device, right? I think there's a recognition that, you know, all of your AI going through the cloud and all of the data always going to somebody else's data center is not advantageous from a business perspective, from a user experience perspective, from a privacy perspective. And so I'm encouraged that even the, you know, the largest of companies, they're, you know, for whatever their motivation is, but that's already on its way into consumers.

(Joel Beasley at 00:12:36) Yeah. Yeah. That's good. And that's what I want. I want privacy. It's weird because you can read emails and you can read that, but the way that we use LLMs or I'll back up. The way I use LLMs, haven't done research on how other people are doing it. But I talk to it, and I share ideas with it, and I ask it questions, and I have it help me with certain things. And these conversations are not like keyword search terms. It's very obvious what I'm thinking and what I'm trying to do. And the biggest frustration that I have, one, privacy. And two is they've built morals on top of the trained data.

(Zach Goldberg at 00:13:17) Yeah. Yeah.

(Joel Beasley at 00:13:19) Yeah. That bothers me.

(Zach Goldberg at 00:13:20) Opinionated, somebody else's opinionated guardrails. And, yeah. You know, I have not myself done the research to see to what extent they open, but how they're defining those guardrails. I think there's some, I have a peripheral sense that there is some openness and transparency there, but certainly, you know, I'm a relative, I keep up with tech news. I'd like to think I'm relatively informed on the space. I'm not an expert, but relatively informed. I have no idea. I don't actually, I haven't seen it. I haven't read it. It's not talked about certainly in the popular press, this idea of safety. And then how do we define safety and communicate it to the public at large? Like, it may be that Facebook and Google and OpenAI have these huge safety teams and maybe they're doing all these wonderful things, but I don't think it, you and I and the rest of the world really believes it until there's a really concerted communication effort. And we have a common language for describing these guardrails and understanding, you know, different cultures probably want different things, and people might want choice. And so that's yeah. Fingers crossed to three more years, right?

(Joel Beasley at 00:14:29) Well, I think Elon's giving us a masterclass in it right now, getting attacked by all these countries, and he's falling back to laws of the country. Right? He's like, we're trying to just adhere to the laws of the country the best we can if something's illegal or just frowned upon. Those are two very different things.

(Zach Goldberg at 00:14:47) Yeah. Yeah. I don't envy anybody on Twitter's or X's data privacy and data content moderation team. That is a challenging role.

(Joel Beasley at 00:14:58) Transparency, though, is I think what you were getting at. Yeah. I like it. I would like the safety to be overcommunicative. I want them to be exceptionally intentional. I want them to go over and above and beyond to explain why they make certain decisions and how things work versus it being like a black box.

(Zach Goldberg at 00:15:17) Yeah. The challenge, I think, is, you know, the general public. Yeah, I'm confident that whatever safety mechanisms are in place involve a ton of nuance. These guardrails are probably not just as simple as, you know, five or six English sentences. It's almost certainly an elaborate system. And how do you explain that? What is the language we use to describe these rules, these opinions, and who is the audience that has to understand that? And then how do we abstract that so, you know, an article in the New York Times that my wife or my mother-in-law will read, who are not technical people, how they would understand it? And is there a missing certification? We have to delegate trust to somebody here so that the public at large can feel that, yes, there's transparency and somebody else that I trust has went through the nuance and that it's good and that it meets whatever criteria I'm expecting.

(Joel Beasley at 00:16:10) Yeah. And there are some, so I've run into this. I ran into this recently. I was curious. I was asking it to generate me a list of different technology leaders in different spaces, and I was giving it very specific criteria. And I noticed that it gave me weird results in the sense that I know 80% of this industry is male just by demographics, not personal opinion, but it's more than 80%. It's demographically, that's just the way it is. Don't be mad at me for saying the truth. But it was giving me, like, 50/50. Right?

(Zach Goldberg at 00:16:46) And I—

(Joel Beasley at 00:16:46) was like, well, that's odd. And so I asked it. I was like, you know, is this—do you have anything? And it's like, yeah. And I said, okay. It said that it was modifying the results for me, and I saved the conversation. So it's like, yeah, we're modifying the results so they're equitable and all this stuff. I'm like, great. So now you're not giving me the best data. You're giving me the best data then modified by some other value system. Right? And that's frustrating because I wasn't involved in that decision. Right? And so it's like, if that's happening and that specific, how many others? Where is the list? Does OpenAI have a list of value systems that it's embedded into the thing? No, they don't. You just run into them and have to figure them out.

(Joel Beasley at 00:17:28) And so for me, that particularly is frustrating, and it's going to be an interesting problem to solve because there are many different cultures on this planet with many different products. They're not static.

(Zach Goldberg at 00:17:39) What we think is okay now, you know, the idea of misrepresenting statistics for a particular value, maybe that changes in five years. Yeah. Who decides at what point it changes and when we want to go back and update these things? And by the way, do I get updated that my results are now different?

(Joel Beasley at 00:17:57) Yeah. Are you going to tell—

(Zach Goldberg at 00:17:57) me that the bot has changed? It's, yeah, a challenging problem.

(Joel Beasley at 00:18:03) Yeah. But if you ask it for the most popular of something and it's got data to know how many followers each thing has and it diverts from that measure of popularity to be inclusive, that's frustrating for me. And I'd rather just give me the best results. And the other thing that was interesting was I got it to get me what I wanted. Do you know how I did it?

(Zach Goldberg at 00:18:25) Some sort of clever prompt hacking, I'm sure.

(Joel Beasley at 00:18:28) Yeah. No. So I just said that I was, like, going to a men's only retreat and needed, like, examples of men. Yeah. But, yeah, there are certain religions and cultures where the women don't do business. It's men only. And that's not in my religion or culture, but that exists out there in the world. So, like, whose culture is going to win? Is it the—and it's probably the culture of the people that make the tools. Right?

(Zach Goldberg at 00:18:55) Yes. Which is unfortunately commonly California. Right? And the values and the demographics represented therein. And, I mean, I'm in California, so I'm just as guilty. But it is, we are not there yet. With any luck, five or ten more years, we'll have invented another layer of abstraction here to provide better systems. But you know, there's also a lot of good stuff that comes out of this. Yeah. I work with people every day who are learning things, you know, facts, helping with programming, things like that. So I think it is important to be critical and hold accountable the cutting edge so we're developing in a good direction, but also appreciative of the value that's being offered even today.

(Joel Beasley at 00:19:40) Oh, it's so crazy cool. And I'm definitely being nitpicky on this amazing technology. I just think that these are—and we'll kind of wrap it up on this unless you have anything to add. Your point that you made of transparency, I think that's the solution to everything. I'm okay with the fact that they add something on top of their model to make results more equitable. Okay. That's fine. I know that. But I would like that to be disclosed. I would like it to be openly disclosed on, hey, here's our manifest for how we modify the base trained algorithm. Here's all the values we've—and then that lets me, as a consumer, choose. Now I can look at Claude, I can look at GPT, and I can now weigh this as a vector. I can, you know, say, yeah, I'm going to sacrifice some privacy because that one doesn't have any value systems on top of the data, and so I'll use that one. You know?

(Zach Goldberg at 00:20:33) Yeah. I think just my last thought on the subject—I completely agree with you. I like the word you used, manifest. There should be a manifest that in a relatively somewhat technical way describes what these guardrails are or these modifications. And I don't think we can trust that each of the model creators will use the same manifest system. And I think that is actually important, that the language we use to describe the models across models has to be the same and so you can compare them equitably. And then somebody else could, you know, sort of bundle that into a very layperson-friendly description. Right? But I think if we're describing each of the models using different systems, then that hinders our ability to translate these capabilities and these systems for people who are not themselves machine learning experts.

(Joel Beasley at 00:21:24) We'll call it the Manifest Transparency Act.

(Zach Goldberg at 00:21:27) Right? There you go.

(Joel Beasley at 00:21:28) Because that's what consumer protections are for. Right?

(Zach Goldberg at 00:21:34) Absolutely.

(Joel Beasley at 00:21:34) So alright. Let's talk about your book. I want to talk more about your book. I looked through it. I saw a main theme was recognizing that every choice has trade-offs. We could tie that back to what we just talked about. But how do we, as leaders, know that we're making the best decision when there's an infinite number of decisions to make and trade-offs for each one?

(Zach Goldberg at 00:21:53) Yeah. I think—well, I was on a call this morning with an engineering leader. They were asking me a question, and the way they framed the question was, I can do A or I can do B. Right? And my response was, well, actually, I think what you're describing is a spectrum where A is 0% and B is 100%. But as a leader, you have a choice. There's a slider, actually. How much effort you put into this problem could be 20%. You could choose 40%. Right? And those middle grounds involve some sort of compromise. And generally, what we're talking about here is how much tech debt am I willing to accumulate by making this decision? Right? The answer in a little bit more detail on the conversation this morning was, do I do nothing about this debt or do I solve the debt entirely? And I think my argument was, well, you could choose a middle ground where you solve some of the debt, but you just accept that you're going to accumulate a little bit, and at some point, that will need to get addressed. And I think, zooming back out to your question, the idea of trade-offs and the value for technical leaders and what I help people do is it is really enlightening and it helps you make better decisions to have the perspective of enumerating what is the trade-off we're talking about. Right? If you can really explicitly say or write down, these are the consequences of option one and these are the consequences of option two, is there a gray area between them? And just the act of enumerating that and saying it out loud gives you a lot of clarity and gives the team clarity. And if it comes down to a point where, you know, the leader and the team disagree, at least the leader can say, you know, for reasons X, Y, and Z, I really value this element of the trade-off, and so that's why I think this direction is the right one for the team. And so it's the process of sort of jumping back or stepping back and saying, well, before we make the decision, let's be really clear about what is the framework by which we're evaluating the trade-offs. What are the consequences of each of the options? What are the benefits of each of the options? And this doesn't have to be—you're not writing a 50-page paper every time. Sometimes it's just a five-minute conversation and, you know, you go through that. And I think that, you know, it's the decision-making process in general that I'm trying to get people into a very pragmatic, rational view about these decisions.

(Joel Beasley at 00:24:11) Yeah. I like that. I went to see Jordan Peterson back in March. I got to go to one of his talks and meet him after the conference. And what he talked about was the different probabilities in front of you and the more possible paths and related to stress. And one way to solve that is to write it out. Just sit down. Like, if you have anxiety, sit down and just dump your thoughts into a log file.

(Zach Goldberg at 00:24:39) Yes. Absolutely.

(Joel Beasley at 00:24:40) For engineer friends, just dump them out there. And then as you notice, as you start watching it come out on the paper or onto your Evernote, whatever you may use, it starts to take a different shape, and you can kind of move stuff around. You can look at it outside of yourself and it gives you this different perspective versus them all just streaming in your mind.

(Zach Goldberg at 00:25:00) Yeah. It takes active energy. Again, we talked about the brain being a muscle. You have to actually actively exercise that muscle to keep thoughts in there. Right? And to keep things organized in your head takes energy. And the act of writing it down sort of releases you of that responsibility of maintaining that energy. So you sort of get it off your plate, so to speak. And there's a reason why, you know, one of the most—I don't want to quote statistics here—but a common tactic in talk therapy is start a journal. Right? It's the same process. You have all these ideas in your head, things that are causing you stress and anxiety. Putting it down on paper allows you to see it more clearly because your brain can then focus on looking at the problem as opposed to maintaining and organizing the problem.

(Joel Beasley at 00:25:43) Yeah. I wait till I'm going to explode and then write it out.

(Zach Goldberg at 00:25:46) You must be lazy like me.

(Joel Beasley at 00:25:48) Yeah. I would love to say I have a daily journaling habit. Every night before I go to bed, I get out my feather pen and write a letter. No. I would love to do that. It's idyllic in my head. But the reality is, is I—that anxiety trigger flips, and I say, oh, I need to write it down and get my thoughts out, and then I'll be able to decide on this, you know, much easier.

(Zach Goldberg at 00:26:10) Yeah. And I think, you know, as a leader in technology, you're held to a higher standard. Right? Because now you're making decisions for other people, and you're shepherding decisions perhaps for a larger team. And so having a discipline and instilling a process within the team so the team as a whole can make good decisions becomes really important. So for my own personal life, I'm kind of a lot like you. I write something down in my own personal life only when it's really critical. But at work, I've made a habit and I hold my team accountable to a higher standard, which is when there are decisions that have impact that matters, it's going to cost resources, it'll cost time, whatever it is, we always have this type of document. Right? And they don't have to be 20-page documents. It could be something that's three paragraphs. But there is a discipline in the workplace for making good decisions and documenting some of the rationale behind those decisions.

(Joel Beasley at 00:27:00) Yeah. One thing that's helped me—I don't know, we're just sharing here. One thing that's helped me in the past year is I made a spreadsheet with the system, the business systems that are important to the business, like things that have to happen. And so what I noticed—I've been running this business for ten years. What I noticed is I was always just thinking about, like, oh, where are we at for payroll? What's going on with clients? Have they all paid? Like, all these different things, you know, is the marketing system running? Are new meetings being booked? Are proposals going out? And so, like, this would just always run through my mind. And when I started this one sheet, I just said, okay, here's the five systems that actually matter, that actually make the money so we can make the paychecks. And I just simplified it down to the least amount of things that have to function properly. And then I set a time in my calendar every week to sit down and review those, and I do whatever I need to do when I see them right away. And for me, that just made my life so much easier.

(Zach Goldberg at 00:28:05) Yes. Absolutely. When I'm coaching people, I call these machines. Like, your job as a leader of an organization, whether you're CEO, CTO, director of engineering, whatever it is, if you're responsible for business process and other people, at the end of the day, problems are going to happen. Your business has to do stuff, and your job is to sort of zoom out a little bit and see the pattern. Is this repeating? Is this a thing that's going to happen often? And then put either the documentation or the people or the process or the technology or the automation, whatever, in place so it becomes a little machine that runs on a loop. Right? Businesses are just made of these little loops. You have a marketing loop, a sales loop, a payroll loop, of course, then your product development loop, whatever it is, all of these little loops, these little machines. And then you as an operator, you're adding a little bit of oil here and you're optimizing here and there. You're trying to create this system where each of these loops sort of runs automatically and in a way that's continuously improving and generating value.

(Joel Beasley at 00:29:06) You have to watch the loops, though, because I noticed if you take your eyes off the loops, if you remove the monitoring of the loops, even with qualified, talented people, the loops just decay.

(Zach Goldberg at 00:29:17) It's interesting. I think there's a number of philosophies on why that happens and ways to prevent it. Quoting the book that I read most recently, of course. You know, Netflix has this freedom and responsibility element of their culture. And I think if you were to ask Reid, his take would be if people are informed and they know the direction the ship is trying to go and they have the context, then they can self-correct when the loops go awry.

(Zach Goldberg at 00:29:45) And it's only when you have empowered people who lack context where you end up going in the wrong direction or those folks who weren't qualified to begin with. So assuming they are qualified and they are high performing and they have context, then you should be able to trust them to make good decisions and make sure that the ship is pointed in the right direction and that the loops are operating. I think the challenge often is what does it mean to hire people with good judgment? And I think people underestimate quite significantly the amount of effort it takes to make sure the entire organization has that context, or at least the context is shared appropriately within an organization. Over-communication is really hard.

(Zach Goldberg at 00:30:22) I tell people all the time, it is really easy to under-communicate. It happens every single day by accident without even thinking about it. But to actually communicate too much is basically impossible as a leader. You will end up saying the same thing over and over and over again. And from the perspective of your employees, some of them still haven't heard it.

(Zach Goldberg at 00:30:40) And so it doesn't hurt to try really hard to make sure these messages—especially important business context and strategy—you've got to say it and you've got to write it and you've got to present it. And it's just going to be a regular motion.

(Joel Beasley at 00:30:54) When I was just a software engineer before I had managed people and all of that, I personally would get so frustrated if I had to repeat myself. I'd be like, I said the words, I did the effort, I made the noises with my monkey mouth, it's been transmitted. Who would have thought that the thing I hated most in life—I ended up becoming a public speaker, which is essentially going around city to city giving the same talk as if you gave it the first time. So I've took that thing I hate and turned it into something that benefits me. But then I realized as I started going about this journey of working with other people and managing them and growing and building businesses that most things are just you being consistent. Do you have kids?

(Zach Goldberg at 00:31:39) My first in two months.

(Joel Beasley at 00:31:41) Oh, congratulations. Yeah.

(Zach Goldberg at 00:31:43) Thank you so much. Yeah, timely question.

(Joel Beasley at 00:31:45) Yes. I've got three under the age of 10.

(Zach Goldberg at 00:31:48) Okay. How do you have time?

(Joel Beasley at 00:31:50) Yeah. We homeschool too, so they're on the property. We have a separate building for the studio, but I go back and forth throughout the day.

(Zach Goldberg at 00:31:57) Oh, that's awesome.

(Joel Beasley at 00:31:58) Yeah. Yeah. But most of parenting and most of leadership is just being consistent, maintaining that vision and just restating it. I found this trick so strange. When I'm working with a lot of freelancers, I'll say what I want, and then they'll say a bunch of stuff back to me that's not exactly what I said. It's like going to a restaurant and ordering a chicken sandwich, and then them just saying all of this information back to you. It's like, just say that you're going to just confirm the chicken sandwich.

(Joel Beasley at 00:32:31) Yeah. This is transactional, freelancing. This is what I want. And so what I found is I used to get upset about that even after having managed people. But then I realized that—who was it? Somebody in my life told me, they're like, oh, you realize that when somebody asks you a question, you don't have to actually respond with the answer. And another tool is if you ask somebody to do something and they don't give it back to you exactly, you don't have to explain it in a different way. You can just say the same sentence back to them again. And I was like, no way. And it works so well.

(Joel Beasley at 00:33:07) So I'll say something, and then, let's say email marketing, working with an email marketer. I'm like, hey, this is what I need. They come back to me with all of these crazy ideas that are not what I asked for. And I'm like, nope. This is what I need. And then, this is what I need. And then finally, I'll get to the point where I'll say, is this something that you are able to do for me? Or I'll get it to a point where it's like, you can either achieve this or I can just go find somebody else to achieve it because this is the thing that needs to be achieved is immovable. That has to be achieved.

(Joel Beasley at 00:33:39) And so I found that is fairly successful—just restating exactly what you want versus my—and I believe if it's my default reaction, a lot of people will identify with it—getting frustrated and explaining it and thinking that they're stupid. No. They're not. You just have to be—they haven't learned the skill of telling you exactly what you asked for back and then doing that.

(Zach Goldberg at 00:34:02) Yeah. I find language the funniest thing. You know, you and I are speaking English today, right? Nobody designed English to be good at communicating information, right? English is a language that's evolved over hundreds and thousands of years from all sorts of roots and influences in different countries. It's an organic thing. And so perhaps we shouldn't be surprised that this unengineered, accidental tool that we use to communicate ideas from my brain to your brain is imperfect and leaves room for ambiguity and is biased by how your brain perceives words that maybe I have a different perspective on. And yet, despite this, we have this expectation that I say something once and it's always understood perfectly in the same form in your brain as it took in my brain. But that's never the case. And it's not your fault and it's not my fault. Language is just a hard problem. And I think the expectation needs to be, especially as a leader of other people where you are responsible for getting ideas across in a way that people will then act on that affects your business, to test how you're communicating things, to iterate on the language, to do it multiple times. It is an effortful process. It is not like you and I talking right now.

(Zach Goldberg at 00:35:22) Right? When there's something important in the business that everybody needs to know, I'll talk to a peer, we'll talk about just like this a couple of times and figure out how do we create the clearest version of this so people actually understand what we're talking about. And then we'll say it several times in many different formats.

(Joel Beasley at 00:35:41) I was doing an interview with Brendan, who's the CTO for Dave Ramsey's portfolio of companies.

(Zach Goldberg at 00:35:47) Mhmm.

(Joel Beasley at 00:35:48) He's a personal finance guy. And he told a story that went pretty viral on Instagram. Got almost 700,000 views, the clip of it. But he basically said for Chick-fil-A to roll out the "my pleasure" thing took six years. Six years from the time the CEO had experienced it, I think at the Hyatt or something. He had experienced it in his personal life, some other brand doing this, to the time where that was actually rolled out to the entire organization at large took six years.

(Zach Goldberg at 00:36:21) Wow.

(Joel Beasley at 00:36:22) You'd think it would be a memo. You would think it would be like, hey, this is what we do now. And it's like, yes. But you have to just be so repetitive year after year after year, new hire after new hire after new hire to get it to where it just becomes something that the company is known for. And now we all know them for that, don't we?

(Zach Goldberg at 00:36:41) Yeah. Absolutely. Yeah. And, I mean, I think it speaks to the challenges of communication, of company culture, of process. You're talking about onboarding, documentation, training, manuals, performance reviews, right? If you forget to say it so many times, I'm sure that there's some manager who's overseeing and there's a system to correct and all of the different pieces that have to be put into place at that scale, right? And to have that level of consistency. It's, no wonder people write whole books on this.

(Joel Beasley at 00:37:10) That's one KPI. That's why I'm an entrepreneur. That's one KPI I don't want to be measured on. Joel, you didn't say "my pleasure." 32%. 32%. You need to be at 110%. Sometimes you need to say it twice. No.

(Zach Goldberg at 00:37:26) I think it's actually—it brings up an interesting point. This is literally what I was thinking about in the shower this morning.

(Joel Beasley at 00:37:34) Right now, all the listeners are now in the shower with Zach. Proceed.

(Zach Goldberg at 00:37:39) Take yourself to wherever it is that you have good thoughts and clever thoughts. And I was thinking businesses have different motivations, right? What does every business want, right? Create shareholder value, right? Get rid of the MBA speak. They want to make money. Right? But businesses have different motions and loops by which they do the making of the money. Right? If you're Apple, your ability to continue to be profitable depends on making good consumer products. Right? If you are a hospital, your ability to continue making money means you don't kill people. Right? And quite the opposite, hopefully you're helping them. Very different KPIs. Competitive consumer products, keep people alive. And what sort of activities are people doing on a daily basis to achieve those goals? Right? Well, at Apple, you want to try new ideas and you want to be innovative and you want to explore different materials and different fancy marketing messages or whatever. If you're a hospital, you want a very clear procedure for how to do that surgery and you want every surgeon to follow that checklist perfectly, right? You don't want to accidentally put the scalpel in the wrong place, right? So one is all about taking risks, right, trying new things, and the other is mitigating and managing risks. And so, you know, I think I agree with you being an entrepreneur, 99% of entrepreneurs find themselves in that risk-seeking scenario.

(Joel Beasley at 00:38:56) Oh, I like this.

(Zach Goldberg at 00:38:56) Whereas if you're Chick-fil-A, right, you have to preserve your brand, you have to preserve—and the way you do that is consistent quality, consistent service, consistent look and feel. And if you deviate from those things, then that is actually what puts your business in jeopardy. And so there's this very different mindset where your business activity is preserving and limiting damage potential versus actively seeking change. And I think there's different personality types and different people operate better in different regimes. And I think you and I have perhaps both learned that we are much more in the enjoy risk opportunities.

(Joel Beasley at 00:39:35) You know where I learned that? It was a textbook lesson in the pandemic. That's when I started researching this quite a bit because I was like, I'm not wearing a mask. I'm going to be fine. It's going to be cool. And I realized that brought up—I don't get angry at people who want to do what they want to do. But what it did was it put a visual to something I haven't thought about. I've just kind of been going about my life being an entrepreneur. I haven't sat down and had a conversation with myself about where do I sit on the risk spectrum of homo sapiens. I haven't thought about that. And then when this happened, I was like, oh my goodness, I'm on the high tolerance of risk side of the spectrum. And I'm kind to everybody. Right? But that self-awareness was unique and different.

(Zach Goldberg at 00:40:23) Yeah. What's funny is I think you've talked to a lot of investors and experienced operators. You have to be the kind of person that likes risk to be an entrepreneur, but your job as an entrepreneur is actually not to seek risk.

(Joel Beasley at 00:40:37) The startup is the risk. And then your goal is to reduce that risk as much as possible to make it a successful company.

(Zach Goldberg at 00:40:42) Yeah. Elon says it great. He's like, just step one, you're in a state of failure. When you file the LLC paperwork or corporate paperwork, you are out of business. You don't have income. You don't have—so it's like you're in the 100% risk. TDD, test-driven development. Right? Test fails immediately, and then you've got to get it to succeed. And so I've taken that perspective, and it's been incredibly useful.

(Joel Beasley at 00:41:06) And humbling as well. Right? Oh, very humbling. Yeah. It matures you really fast, especially the first startup that you do when you go broke or whatever. You're like, oh, I never want to be here again. How do I never get here again? And how do I maintain that creative risk? And then you just start looking at other people in the market and seeing what the people who've come before you have done, and you take their little sayings like Bezos, I make bets, but I don't make bet-the-company bets.

(Zach Goldberg at 00:41:33) There you go.

(Joel Beasley at 00:41:35) That's something I can actually build on top of. I can use that in my decision making.

(Zach Goldberg at 00:41:41) Yeah. Absolutely. I like that. I make bets, but I don't make bet-the-company kind of bets.

(Joel Beasley at 00:41:46) Yeah. Yeah. He's made a—he's lost a lot of money and he's done a lot of stupid things, but he never did a move that was, like, I'm betting my company on this.

(Zach Goldberg at 00:41:54) Yeah. I use the Bezos' one-way and two-way doors, type one, type two decisions all the time. Right? An irreversible decision for the business, you want to be pretty thoughtful about that.

(Joel Beasley at 00:42:08) Yeah. What's the best misunderstood piece of advice that you've fielded with people?

(Zach Goldberg at 00:42:16) Misunderstood piece of advice.

(Joel Beasley at 00:42:18) Yeah. Like, they think of it one way, and they're like, that's what I think it means, and they try to implement it. And it's like, oh, no.

(Zach Goldberg at 00:42:23) Yeah. It's speed at startups.

(Joel Beasley at 00:42:25) Okay. Tell me more.

(Zach Goldberg at 00:42:27) Not every engineering leader—I don't want to say everyone.

(Joel Beasley at 00:42:31) All of you, every single person listening. Yeah. What are they doing?

(Zach Goldberg at 00:42:37) Let's describe this from first principles. There's a lot of pressure to move fast at a startup, right? You want to mitigate risk, you want to build a product really quickly, demonstrate product-market fit really quickly. And so engineering leaders, CTOs at startups, everybody says, I have to go as fast as possible. And I think there is a misunderstanding that going as fast as possible doesn't necessarily mean skipping all of the good engineering practices that ultimately are what allows you to have engineering velocity long term.

(Zach Goldberg at 00:43:06) And so I talk to engineering leaders who are like, we just write code, we don't bother with tests. And I will say there are some circumstances where maybe that makes sense. If you know you're throwing your code away in three months and it's just a prototype, by all means. But by and large, tests enable and automation enables velocity. Right? If you're going to make a change to your code and you don't know if the code's going to work and you have to manually test it after every code change, that slows you down. And the act of writing the test actually makes you faster. And so there is, relatively commonly I find, teams are skipping some of these best practices. They're not writing specs, they're not doing tests, they're not thinking about debt and actively having strategies in place to manage debt. And then they come to me and they're like, why is my dev team so slow?

(Zach Goldberg at 00:43:50) And then we come back to first principles. Well, actually, what you want is good foundations for your team, and that's what ultimately enables long-term velocity and your ability to execute.

(Joel Beasley at 00:44:00) I've never met anybody who enjoys writing tests. There's some, like, I enjoy writing tests, but I've never met—and I wasn't always that way. What happened was I was just sitting there refreshing terminal constantly. I was doing it in such an inefficient way when I met somebody who was like, "You're doing it that way?" And then I saw how they were doing it. I'm like, "Wait, you just write it once?"

(Joel Beasley at 00:44:21) And then it runs every single time, and I got my testing suite. Obviously, the experience I have in my head right now is very specific to the stack I was using.

(Zach Goldberg at 00:44:29) Mhmm.

(Joel Beasley at 00:44:29) But once I learned to love—Sandy Metz taught me to love writing tests. She was this author, Ruby author, that wrote examples in relation to bicycles. She always uses bike as an example and would show all these tests. This is like twelve years ago. And following her and reading her content, I was like, "Oh, okay. This is how you love testing."

(Joel Beasley at 00:44:50) And I wrote a blog article called "Sandy Metz Taught Me How to Love..." or something.

(Zach Goldberg at 00:44:56) And it—

(Joel Beasley at 00:44:56) It was like dot dot dot "writing tests." Yeah.

(Zach Goldberg at 00:44:59) Yeah. You're very clever with your headlines.

(Joel Beasley at 00:45:03) So, yeah. But I've never met people that enjoy writing tests that have existed in cultures where they're writing the tests as part of the culture of the company who then go, "I'm gonna go do a startup and I'm gonna just write some throwaway code and not write tests." Once you get that muscle, you're like, "This is powerful, and I'm never gonna deviate from it."

(Zach Goldberg at 00:45:20) I think the failure mode and why people often hate tests is testing has to have a purpose that people understand and feel. Right? If you're writing tests that actually make it easier for you to write code, like the automation is catching bugs and the tests are relatively easy to maintain and they're adding value on a regular basis, then it's pretty natural for people to augment and maintain the tests. If you're writing tests because somewhere someone has decided that you have to have 90% test coverage and you're just rote testing every single if statement with these boring, mocked-out tests which aren't actually helping you, then you really hate writing tests. Right?

(Zach Goldberg at 00:45:57) And so it's really up to the organization to set a standard that your testing provides confidence. And if it's not doing that, then you need to not be doing it or write tests a different way. It's not worth your time otherwise. Right? There has to be value that's being provided.

(Joel Beasley at 00:46:11) Yes, I can agree with that. I can confidently say that if you're writing tests and it's not making you a faster developer and you don't see any benefit, you should definitely consider learning from other people how they write tests. Because when you—I don't know how to describe it, but when you figure it out and it clicks in your head how to write tests for your use case, you're like, "This is amazing."

(Zach Goldberg at 00:46:33) Yeah. And I think there's also the sense of inertia. You'll see somebody else wrote a bunch of tests, and now you're working on a feature and your feature is breaking those tests. And you're four hours in to debugging and trying to understand those tests and tweak them for the modified behavior. Is that actually a good test then?

(Zach Goldberg at 00:46:52) Like, what is the correct thing? Do I spend another four hours trying to understand this test, or do I blow it away and write something else that actually is more updated with the new behavior or whatever has changed in your product? I think there's a real reluctance to change old tests that somebody else wrote in the face of changing product requirements.

(Joel Beasley at 00:47:12) You know how I think about it? It's gonna be real stupid, but it's the truth. It's one of the ways that I helped convince myself to start loving tests. Right? Because you can love—you can love working out. You can make yourself love things that are good for you even though you don't love them. You just say you do, and then it happens over time. I would imagine that I would take a little Keebler elf. You know, those little Keebler elf cookie guys that live in the tree and make the cookies?

(Zach Goldberg at 00:47:35) Yeah.

(Joel Beasley at 00:47:36) I was like, "Alright, I'm gonna take the little Keebler elf, and he's gonna sit there and watch that piece of code and make sure it doesn't change." You know? And that's how I would look at a test. I'm like, "Oh, I'm making another little elf, and this elf is like my little assistant, and he's just gonna watch that and make sure it doesn't change and that everything's good. And every time I make a code change anywhere in the whole project, it's gonna double-check that that thing's still working as it's intended to." And for some reason, whatever stupid way my human brain works, I was like, "Oh, come on, elves. We're gonna go do work today. Let's get going."

(Zach Goldberg at 00:48:06) Yeah.

(Joel Beasley at 00:48:09) Slow test? That's Dopey, bro. What's he doing? But you've—

(Zach Goldberg at 00:48:14) Named all of the tests.

(Joel Beasley at 00:48:17) No. No. Oh, yeah. But you have to find fun ways to keep it interesting because I've got your monkey brain. Right? And you have to keep that thing entertained.

(Zach Goldberg at 00:48:26) Absolutely. Yeah. Yeah. It's—I've you've heard of, I'm sure, the rubber duck debugging?

(Joel Beasley at 00:48:31) No. Tell me about it.

(Zach Goldberg at 00:48:32) Oh, rubber duck debugging? This is great. It's a—there's a Wikipedia article on it. I'm not the inventor of the phrase. But the idea is, you know, talking out loud is a great way to problem-solve. And so you've got a problem, you call a coworker, and in the process of explaining the problem to the coworker, you self-identify the problem. But now you have the unfortunate side effect of you had to disrupt your coworker and they had to listen to you for five minutes. And so I've actually—my wife makes handmade glass. So she made me this glass rubber duck.

(Joel Beasley at 00:49:02) Oh, look at that.

(Zach Goldberg at 00:49:03) So it's this cute little guy. And so rather than talk to a coworker, you talk to a rubber duck and the rubber duck will solve your problems in a couple of minutes is what tends to happen. So—

(Joel Beasley at 00:49:13) Oh, so I feel completely cool about my Keebler elf thing right now.

(Zach Goldberg at 00:49:17) Absolutely. It's official. There's a Wikipedia page for it. You're totally good.

(Joel Beasley at 00:49:19) Maybe that's kind of what I was doing. I was having my little elf there, like, "Hey, what did you change?" No. But I do that. Right when you started describing that, I was like, "Yeah, I'm a verbal processor." I know that about myself, and I've got certain people that I call to process certain things.

(Zach Goldberg at 00:49:38) Yeah. It's funny how, you know, problem-solving is somewhat social. And you picture in your head, you know, Einstein coming up with the theory of relativity in his basement in the 1920s, just in a dark room with an oil lamp writing on pen and paper. But it's not really how innovation happens. Right? The exchange of ideas and communication and being social advances the state of the art of thinking way faster. Right? And so I don't think there should be an expectation that the best computer programmers just sit in a room and independently write code for eight hours a day. Right? There's this fundamental push and tug with other people and identifying solutions that just moves faster and gets better work.

(Joel Beasley at 00:50:18) That's funny. The images in your head, when you think programmer, hacker-type person, you think someone alone. And I spent a lot of time alone writing code. I got better exponentially faster when I got on a team.

(Zach Goldberg at 00:50:31) Mhmm.

(Joel Beasley at 00:50:31) Around other people versus just kind of doing it myself with Stack Overflow or whatever. Yeah.

(Zach Goldberg at 00:50:36) The bottleneck, especially with Copilot nowadays, the bottleneck for writing code isn't the fingers on keyboard typing characters and typing individual lines of code. Right? The bottleneck is understanding the problem and then understanding how to, using the tools in front of us, how do we solve that problem? And to some extent, you know, that process occurs at different scales and different altitudes. And some of that lower altitude, yeah, absolutely, developer just needs to put time into it. But a lot of that problem-solving occurs at a higher level than just the code. Right? How do we plug this widget over here into this other widget over here? Do the APIs exist? What would the design of that API look like? Or is it a common problem, for example? And I think one developer sitting alone for two hours writing that up is probably a lot less productive ultimately than two people for half an hour just talking it through and understanding the problem.

(Joel Beasley at 00:51:24) Absolutely. I want to talk about these different types of CTOs. In your book, you said that there's tech-focused CTOs, there's people-focused, and there's externally focused. Tell me about them.

(Zach Goldberg at 00:51:40) Sure. And so this is partially a description of personality types, and it's partially a description of business needs. And in summary, it speaks to the ambiguity of the phrase chief technology officer and CTO. Right? You have somebody at your company who is your most senior, you know, whatever, and they become your CTO. But that person has a background as presumably a software engineer. And you have many other software engineers. So what are they doing compared to your day-to-day individual contributors? And I think the analysis is from the business's perspective, there's three key types of activities that your most senior technical person can be doing. The most obvious of which is they're driving technology. Right? They are figuring out what is the future of technology at your company. And so this is maybe also often called a chief architect. Right? They're prototyping. They're working with an office of the CTO kind of thing where they're developing the next version of the product, coming up with ideas, doing the thought leadership. And there are for some businesses at some times, that's the most valuable thing for the person to be doing or that's the thing that person's best capable of doing. Then there's the sort of VP of Engineering-style CTO where their expertise is building an organization of people and setting the culture and process of that organization. And during rapid growth phases, this is often the most valuable activity. Right? So hiring really well, putting management training in place, making sure the teams have good tools, that there's good standard processes for how things get done, so on and so forth. Excuse me. And then the third and final kind that I've described is the external-facing, which is really the sales CTO. And this is a case where the company has a viable product, you're in market. But in order to sell that product, for whatever reason, you need the authority of a senior technical figure in the room to either speak to the problem or speak to the roadmap or explain how it works in a way that comes across well for customers. So sort of the sales CTO, so to speak. And I think the key idea is that, generally, any individual person will have different levels of expertise along these three different personas. And it's important for you as a leader to recognize, what does my business need right now? What is my expertise? And if those two things are not aligned, then you need to adjust the circumstances so that somebody whose expertise is aligned with that value is doing that job.

(Zach Goldberg at 00:54:10) And so if you are really a technical architect, but you're in a CTO position where your company really needs a chief technical salesperson, then the advice of the book is hire somebody or find somebody on the team who has that sort of gift of gab, so to speak, and can represent the company well externally and speak to the technicals, right, rather than do it yourself. It's better to have somebody who's in their zone of genius doing that highly valuable activity. And I think that the challenge is it's all under a single job title. And so it's unclear if you just say, "I'm hiring a CTO," or "We need a CTO," what is the actual minute-by-minute, day-to-day value that you're expecting that person to generate? And the title alone doesn't communicate that. And so there needs to sort of be a deeper exploration of where is the value generated and where are you best suited for generating the value?

(Joel Beasley at 00:54:57) Beautiful advice. And that's what I do with interviews, in the sense, like, if I was interviewing to be a CTO.

(Zach Goldberg at 00:55:05) Mhmm.

(Joel Beasley at 00:55:05) You know, they'll put a job post out or whatnot of you need to know these stacks and these areas and these similar business things. But at the end of the day, whether HR wrote that up based off of research they did on what they think a good CTO would be, you have to figure out why is that CEO and COO asking for a CTO? What is the thing that they think the CTO will bring? And then is that my strength? Because you're not gonna be happy if you aren't valuable there. You know? If they need that salesperson and you're super shy, it's—you're not gonna be happy, and they're not gonna be happy because they're gonna expect you to be hyper-social and you're not.

(Zach Goldberg at 00:55:46) Yeah. Yeah. And I think most importantly, I just want to underline, that's not a judgment. It doesn't make you a bad CTO. Just not a good fit for that particular company. Right? You could be a great CTO elsewhere where it does need, you know, people and other skills.

(Joel Beasley at 00:55:58) And I encourage people to be more picky about how they spend their time versus the—obviously, money is a huge component. Right? But you're gonna be miserable if you're making an extra $50K, but you're doing something you hate versus making $50K less and doing something that's immediately in your wheelhouse, something that you can double down on and really develop a better reputation in the industry and in your career because you're on fire about it. That's—I see a lot of people that haven't figured that out yet in the job market.

(Zach Goldberg at 00:56:34) And that's where your career growth is gonna come from as well. Like, you're gonna learn more if you're energized and excited by the actual work that you're doing. And people will see that as well, and you're gonna have a bigger impact. So all these positive effects from aligning your zone of genius with where the value is generated.

(Joel Beasley at 00:56:50) Yeah. Well, Zach, this has been amazing. I think I just made a new friend. Dude, thank you so much for doing this.

(Zach Goldberg at 00:56:55) Thanks for having me. This is super fun.

(Joel Beasley at 00:56:57) Thank you so much for listening. And if you found this episode useful, please share it with a friend or colleague who you think would get value from it. And if you have topics that you would like to hear discussed on the podcast, either add me on LinkedIn or send me an email, [email protected]. Every time I get an email or LinkedIn message, it absolutely makes my day and inspires me to keep going.