Welcome And Masterclass Note

SPEAKER_00 0:22

Getting really good at just switching tools and adapting and learning it. Yeah. That's the number one skill set right now. It's not, oh, I'm really good at this one tool because I said sleepfrogging.

SPEAKER_01 0:34

Once again, welcome to the Digital Customer Experience Podcast with me, Alex Turkovich. So glad you could join us here today and every week as we explore how digital can help enhance the customer and employee experience. My goal is to share what my guests and I have learned over the years so that you can get the insights that you need to evolve your own digital programs. If you'd like more info, need to get in touch, or sign up for the weekly companion newsletter that has additional articles and resources in it, go to digitalcustomer success.com for now. Let's get started. Hello and welcome back to the Digital CX Podcast episode 106. And I am pleased to welcome back this week Eric Mystery, who, since he was on last, maybe a year or two ago, has joined Zapier. And if you listen to the last Eric Mystery show, um he's very much involved in customer education. But he's also been just deeply entrenched in artificial intelligence basically from day one, and has a lot of kind of that background of the evolution of AI and continues to play, especially in his new role at Zapier, where he's basically combining all of his worlds from customer education to playing around with AI and uh just really helping bolster you know Zapier customers' ability to use the tool sure, but then also to use the ecosystem of tools around them. Pretty fascinating conversation if you're interested in where AI is, but specifically in the lens of customer experience and customer education. This is gonna be a great episode for you. One quick housekeeping note uh registration is still open for the first cohort of the Digital CX masterclass. It's a six-week cohort program that I'm gonna be teaching all about Digital CX from inception and strategy all the way through to execution and maintenance and all that kind of stuff. We talk about org design, we talk about tools, we talk about artificial intelligence, we talk about all kinds of fun stuff. So um check it out. Digital Customer Success forward slash masterclass. All right, let's get into the show with Eric Mystery. Eric, you are, I think, one of I want to say seven or eight people that have returned to the show. So you're in an elite club.

SPEAKER_00 3:04

I was gonna say that's uh how it company right there.

SPEAKER_01 3:07

So welcome

Eric’s Career Shift To Zapier

SPEAKER_01 3:07

back.

SPEAKER_00 3:08

Got an invite back, and yeah, you weren't like, wow, that guy said some weird stuff. Yeah, no. Never gonna talk to him again.

SPEAKER_01 3:14

Well, it's interesting because we've been talking on and off here and there, and um, you've had a lot of changes over the past year or so, and so I figured we'd talk about that a little bit. But when we first chatted, you were part of the Content Square team. There's been a few Content Square people on, and uh, you know, deeply focused on customer education and stuff like that, but you had already solidly um explored and been part of the AI fund that was emerging like two years ago when you were on. And so fast forward to today, do you want to talk a little bit about what your journey has been?

SPEAKER_00 3:50

Yeah, it it has certainly been a tumultuous journey. Uh yeah, just last Friday was my uh my 34th birthday. So I was like, oh, what has happened over the past year? And man, it's been a lot. So uh Content Square, my role and team kind of got eliminated as kind of post-merger things happen. Um and then from there I went briefly to a company called SmartCat, which was AI translation to help launch their customer education function, basically. Right. Uh and then in September I got invited by Zapier to basically do my dream job, which is AI transformation, yeah, but also through a teaching and learning lens. Like, how do we actually move people from wanting to be AI transformed to actually AI transformed? So that's been an incredible journey in like getting to basically work at my dream company into Zapier. Zapier.

SPEAKER_01 4:45

Yeah. Um, it I was so excited when I first saw that you were moving over there. I was like, yeah, that yeah, you know, the the shoe fits.

SPEAKER_00 4:54

Yeah, I think someone said, Oh, the kid got led into the candy shop, and that was pretty much accurate.

SPEAKER_01 5:00

Um so talk to me a little bit about your role specifically.

SPEAKER_00 5:05

Yeah, so day to day I'm doing a lot of different things. Um, one part of my job is working directly with clients, like actually sitting in with their company, talking to them and understanding like

AI Transformation Starts With Culture

SPEAKER_00 5:18

one, what do you want to do with AI? And two, how do we weave that into your culture? Because I think the biggest thing I've seen a lot of people, especially on LinkedIn, get wrong is it's not about the technology. Like this whole AI revolution is about culture change, it's about getting people comfortable with it. You're always gonna have people like early adopters, like, I'm always gonna look for that weird tool. Yeah. But one, you're not gonna have a company of a bunch of people like me, two, you don't want a company of a bunch of people like me. You need to get everyone on board, everyone using the same shared language, the same shared goals. Like, how should someone in 2026 be using AI tools? Um, yeah, and that's a lot of what I explored day today.

SPEAKER_01 6:06

That's cool. So your remit is essentially AI education for Zapier customers. But then there's, I'm guessing it's a big big part of this is just thought leadership and and being on shows like this and stuff.

SPEAKER_00 6:20

Well, yeah, and it's it's also just you know, internally, we're really trying to push that envelope. Like AI fluency is a major tenet of like how does Zapier work? Um, because if we're gonna be building those products, if we're gonna be saying, you know, we're AI transformed, we want to be on that cutting edge. And so those of us who are early adopters who are willing to experiment, then we also get brought in to show the rest of the folks at Zapier who like, you know, they might be pushing the edge in some areas, but may not know exactly how to use these AI tools. We're all teaching each other, yeah, and also trying to like raise that level. Um, and then also I'm doing a lot of work with getting uh Zapier Academy kind of back up and running. Cool. So yeah, how do we turn, you know, I work individually with companies. Yeah, how do I turn parts of that into something scalable that everyone can use? Because I also think that AI, while it's simultaneously like an equalizer, it's also one of those things where the future is not equally distributed. Yeah, if you have access to the tools and access to advanced models, you can do more. And so I don't want to have like knowledge be the gap that keeps folks from doing with it.

SPEAKER_01 7:28

For sure. Um, I was talking to somebody else today actually about this on a call, and we were talking about the fact that um it almost doesn't matter what if you're in the Gemini

AI Primitives And Switching Tools Fast

SPEAKER_01 7:43

camp or the Claude camp or you know the OpenAI camp or whatever camp you're in, like it kind of does if you're in the minutiae, but really if you're um if you're analyzing a heap load of data or you're just you know trying to automate your life or whatever, all of these tools are just kind of leapfrogging each other. And and so I like your approach of like you're not teaching like this thing, like you're not going hard on Claude or OpenAI or OpenAI or whatever, but you're it's more about the skill involved and the the the way of thinking, sure, prompt, you know, prompt engineering or whatever. Even that's scary to a lot of people, yeah. But it's more about like how are we using this on a day-to-day basis, and what are the habits to get into, and how do you think about your daily routine?

SPEAKER_00 8:35

Yeah, and I'll and a lot more of it is like more transferable than people think. I I think I've switched camps like many times, and yeah, I think day-to-day I switch between models, I switch between products, like I'm using cursor, I'm using cowork, I'm using clawed code, I'm using codecs, I'm also playing like Google's AI Studio. It doesn't really matter what tool you're using in the end, because right now we're still in such a fluxy mode that it is one of those things where if you can understand almost, we call them like the base principles, like the primitives. Like, what are those AI primitives? And they're all gonna call them different things. Like I use cursor, they call a chat an agent, which means it's a thing in all these different products. It's like, okay, I know I'm gonna have a conversation with a model. I know I can bring context in. Some places they call that skills, other time places they call it other things. It's like understanding those building blocks I have available and how I bring them along is the main thing I think about. And also just because things are moving so fast, getting really good at just switching tools and adapting and learning it. Yeah, that's the number one skill set right now. It's not, oh, I'm really good at this one tool because you said sleepfrogging.

SPEAKER_01 9:52

Yeah. And you know, just because you're a master at Claude code doesn't mean, you know, your next client that you advise or your next company is gonna have Claude even, you know, they might they might be in an entirely different data security privacy ecosystem.

SPEAKER_00 10:09

Yeah, and that's the other really interesting thing, seeing all these different companies, because my first like job out of college was working in healthcare software. Oh, yeah. And it's one of those places I'm like, how do you manage that? How do you manage the privacy and you know, dealing with healthcare data, the consistency you need with that data? It's so it's one industry is also another huge factor that's playing into all of this. And you know, some of us jump between industries. If I'm working in the healthcare industry right now, and I'm not in my work able to use these tools by virtue of you know privacy problems that haven't been solved or a company culture where it's like we're gonna value privacy, which I'm pretty happy about as someone who has healthcare data. Sure. Um, you know, where does that put you when you're trying to move to your next role? Like how and where are you supposed to learn these things?

SPEAKER_01 11:00

Yeah. So let's talk about that for a second because um uh a few weeks ago at the customer success summit, we did in Austin, we did um, you know, a couple of different things. One of them was like a live vibe coding session

Learning AI With Privacy Constraints

SPEAKER_01 11:14

where you know we solicited ideas from the audience, you voted on one, uh, we got some feature requests in, and the next day we went to town. And that one of the one of the rules, so to speak, of that exercise was we're creating a tool that doesn't rely on a data source, right? That doesn't rely on being integrated anywhere, that doesn't really rely on like you know specific customer information, mainly because I wanted to demonstrate the fact that you can still do this stuff without having all of the connectors and you know, insight into your CRM and stuff like that. So for folks who are in a restricted environment like that, yeah, what's what what would you say to them in terms of like, you know, I know you can't do this, but do this.

SPEAKER_00 12:06

Yeah, I mean I kind of do this on my own. I was doing this also before I joined Zapier because at Zapier I've basically been given this gift basket of here's every AI tool you could want to use. Yeah. Please go push the limits. Uh before that was basically, hey, you are on your own if you want to experiment on this on your own. Um, the things I like to do when I'm looking at that problem is one, I just give myself little puzzles to solve. So I'm like, how could I make something like this? Like, how can I keep track of my plants? How can I, you know, build a little app for like my kids' daycare schedule? Um, how can I like our gym's schedule is terrible? It's like this random PDF that's impossible to read. How can I build something that just parses that site and makes it like Eric readable? I only care about lab swimming when four lanes is open. Yeah. That's all I need from the website. Yeah. How can I build tools that help me solve those like little puzzles? Yeah. Because then again, it comes back to those core principles. How can I use those tools? Um, and I think every one of these um tools is also starting to put out really good documentation because documentation is a much lighter lift for them these days. And like Anthropic's new academy of things is incredible. Um, and it's like the knowledge is there and it's taking the time to solve it and you know apply it to problems. Uh, my wife also has like a bakery as a business, so I'm like, oh great, perfect use case. Anytime I want to test out like a website design tool, yeah, and she's like, hey, let me let me just try building a version of your site. Like, we haven't pushed any of them live yet. I was gonna say, like, it's it's still Do you need another new website? Oh no, it's yeah, I mean, the old website is still on WordPress.

SPEAKER_02 13:49

Yep.

SPEAKER_00 13:50

Um, and I'm like, I really just gotta swap it over. I've built like three variations on Lovable, they're all better than the site I hand coded. Yeah, like yeah, so much better. Yeah. And I think that's the other thing that's really interesting. It's like, yeah, when I was using Chat GPT, like three when it like the earlier days, the biggest thing everyone pointed out was like, oh, it has flaws here and here and here. And it's like, it does, absolutely we know where this is going. It is going to get better, it's going to get to a point where it's better than the lay person making websites. Yeah. And I think soon it's going to be on par with really good designers. And then when you factor in really good designers using these powerful tools, you're going to hit even more like next level.

SPEAKER_01 14:37

Yeah.

SPEAKER_00 14:37

And that's I think the other really interesting thing in watching this AI space is what happens when you give like a subject matter expert with like really deep expertise and really deep language access to these tools. Like our marketing team is amazing. And just seeing like what happens when you give a marketer really powerful AI tools to solve marketing problems with marketing specific language, yeah. It blows anything I could write out of the water. Like they're building websites. I'm like, wow, I want to go in and do whatever this website is telling me to. I couldn't build that because I don't have that marketing brain expertise.

SPEAKER_01 16:50

Yeah. And it that's where the context comes in, right? Because I mean, anyone can go sign up for a you know, $20 plan of whatever it is. Yeah. Um, but it's not going to do what you want it to do unless you give it that context. So in the in your example of the marketing function, there's probably a lot of stuff that they're working with that has that context of like marketing experience and that language and being able to, you know, walk that walk. Yeah. But I think that's true for just about everything because you're not going to go, you know, to a you know, a blank slate and just expect it to do exactly what you want it to do. There's a little bit of work that goes into, you know, upfront work to provide that context or the skills or whatever we want to call it, you know?

SPEAKER_00 17:39

Yeah. I'm blanking on exactly who wrote about it on LinkedIn. I'll it's my saves, I'll send it to you later, put in the show notes, but it was a brilliant post on how like the work has shifted. Like I have always thought of myself as someone who makes things. Like I make websites, I make automations, like I am a maker. And so the value of my work comes from like the making

From Maker To Director Of Work

SPEAKER_00 18:03

and like the art and time I spend in making. And now and I fully recognize this, most of my work, like 60, 70%, is in the planning and the setup and like crafting the framework to make the work happen. And then the work happens, and then I spend probably 20% of my time reviewing the output of the work. So it's like I assign work almost like a director level, I watch it happen for like 10% of my IC level, and I have 20% like the manager who's checking off. Like, yeah, does this pass? Did my team produce the works that I can trust and push up to the rest of the world? And that shift, I think, is also really uncomfortable for a lot of people. And getting past that personally was very uncomfortable. I am still a maker, but I'm a different kind of maker. And separating the time I spend making stuff from what I produce was a very interesting like brain conundrum. It's like a trust thing, yeah.

SPEAKER_01 19:06

A little bit. It I feel like we're getting into an era of communication. Like I don't know, maker of old, who maybe is I'm not I'm not labeling you as antisocial, but maybe you know, I think of a traditional maker and it's somebody in a workshop building stuff, it's somebody in a home office in their space, in their head, realizing something that they have a vision for. Yeah. And it feels like we're doing that more with communication now, and but we're communicating with you know whatever tool is in front of us, which is just um yeah, it mindfuck. Pardon the pardon the pardon the uh the phrase.

SPEAKER_00 19:51

No, I I've never heard that word before, but uh because like the other thing I'm like, what makes this feel different? Like what is it in this era? And I think where I'm seeing a lot of folks have success is that entrepreneurial mindset, like even if they are not an entrepreneur, that idea of like I'm gonna go and like make something and do something, like I'm going to solve my own problem in a way that others might also want. Um, I think when we look like a cross LinkedIn, it's like, oh, like I've seen like what you're doing, the that whole set of guides, like it's a very entrepreneurial thing. And everyone who I've seen who's kind of done that dive in there and is doing really well, they have an entrepreneurial mindset somewhere in there.

SPEAKER_01 20:38

Yeah. Yeah. It's true. Cause I mean, um, you know, there's so much like niche specificity that we're that's where the opportunity really lies, you know. Um I just I I mean, I often there's so much there's so many layoffs happening right now, yeah, especially among big firms. I mean, I was chatting with somebody from Atlassian the other day, and they're massive layoff there. Um and if you are one of those people and I hope you've been given a lovely severance package, this is like your time to go build stuff. Like, and I think the problem that we all encounter when we go to do this stuff is we get to the keyboard and we're like, you know, like we're really limited by our creativity, and it's forced us to kind of reconnect with this creative uh element of it. And sure, I go through my day now and I'm like, okay, what element, you know, I I almost flag things like this is a thing that I can probably go automate that I won't have to do, won't have to spend as much time on next week or next month or whatever, right? Yeah, so I naturally do that, but we all encounter those situations where either we run into this all the time and we want to do something about it, or everyone is talking about it, like everyone is talking about this one problem, and we're in a place now to where literally anyone can go and solve for those niche problems and productize it. And whether you're in it to make you know a million bucks, or whether you're in it just to be a good member of the community is almost irrelevant. Yeah, we can do that, yeah.

SPEAKER_00 22:30

The the the friction from oh, I have an idea to I I can produce an output is just it's nothing now, yeah. Which I I think is is freeing in a way, but it's also like, oh, I I don't even know where to get started, I can do anything. And whenever I found myself in that, it's just you know, when I was trying to learn first for the first week, I was like, I don't even know where to start. But then the principle I come back to is like just build something because whatever you build first, and this goes for any tool, it's gonna be terrible. You're gonna come back to it like a week later and be like, who was who was the moment that did this? Like, what were they thinking? This is messy, like who wrote this terrible code? Yeah, oh signed by me.

SPEAKER_01 23:17

Yeah, yeah. Uh it's in my GitHub review. Yeah, okay.

SPEAKER_00 23:20

But then the I think the other thing is like the learning loop is accelerated. Like, I started learning cursor about a month ago, and then you know, started building some skills. I went back last week and I was like, let me take what I've learned and build those fresh. Yeah. And it's not like anything else where I'm like, oh, I'm gonna rebuild this from scratch, where I have to put that same time again. It's like, oh, let me take this rough thing and like turn it into a relative diamond. Yeah. Um And that I think is a really interesting piece too. It's like learning doesn't not really cost you anything because you can just learn and iterate, learn and iterate, learn and iterate, because that's that's where you're gonna actually learn. Like, yeah, courses are great. Like take all the courses you can, but build, build them perfectly and come back and fix it.

SPEAKER_01 24:12

Yeah, I totally agree. Interesting. Um I do want to caveat that a little bit. Like, you know, if you're if you're out there and you're going and building a product and you're vibe coding and all that kind of stuff, I mean, unless you we're not at that stage now where you can just like build it in lovable, put it out there, and like be a SaaS founder. I mean, kind of, but yeah, but that's your prototype. I think we you know, people talk about engineering being dead or whatever. It's it's not, it's like it's shifting. You still need security, you still need like all this stuff on the back end to really manage, you know, uh an app and all that kind of stuff. But when it comes to like building yourself tools and building yourself stuff, like that's a no-brainer.

SPEAKER_00 25:06

Yeah, I was gonna say, yeah, a lot of what I build is individual tool tools or team tools, or uh like the rough prototype too is a great thing where it's like, hey, instead of me trying to describe this feature, let me build what it could look like. Because then I think folks are also much better at reacting to something they can see and interact with than something that they just have to like imagine in their head.

SPEAKER_01 25:30

How cool would it be if you um were putting in a feature request for your product team and you like vive-coded a prototype or even some screens.

SPEAKER_00 25:40

I I could tell you how cool it is because that happens regularly. Like I see some a lot of like a lot of folks just do that, and they're like, here's how it could look and feel. Yeah. Um, and you you you see the build teams also react to that, and they are also doing exactly the same thing, too. So um, it is a whole new language of communicating for us. Right. I don't just have to tell, I can show.

SPEAKER_01 26:04

Yeah, totally. Um, one of the things that you were talking about earlier is just like learning stuff, and um I don't know if you've had this same experience, but Notebook LM has been a huge thing for me because if you're not hip to this, like you just put anything and everything about a topic into it, you know, YouTube videos, PDFs,

NotebookLM And Better Self Learning

SPEAKER_01 26:30

docs, copied and pasted something, you give it a bunch of sources, and you can output these really rich learning materials for yourself. Like videos and the infographics and all this kind of stuff. It's unbelievable.

SPEAKER_00 26:46

It it I think it is the most slept-on tool. Yeah.

SPEAKER_01 26:49

Like it should be the one it's the most talked-about and slept-on tool. Like everybody talks about it, but then we quickly forget about it, and so it's still slept on.

SPEAKER_00 26:57

Yeah. And it's one of those things, too, what I what I find myself doing with it is I have a lot of these projects where I'm like trying to bring a ton of stuff together that usually it's coming from a little bit of like ADHD, like disparate fields. Like, oh, I got some stuff from like my NBA field, some stuff from like technology, some stuff from just you know, books and stuff I've read. And I bring it all into there, and I'm like, I know there's threats here connecting this. Let's talk about how this is connected. The other thing I really like with that is when they make the podcasts and when they make the videos, they use really interesting combinations of words that are not in the source material. That I'm like, oh, you took that concept and like gave me the language to talk about it. And when you also start tweaking the outputs of particularly the podcasts, yeah, um, like make them a little bit more argumentative, or try to use really good catchy language in here, you start to see uh things emerge. And I think it's another kind of trend in there where uh voice and like uh voice as a context tool is becoming really uh even more recognized and powerful in this AI space. And those little podcasts are like I I pull out transcripts from them too, and I'm like, how can I bring that into the rest of this? Um because there's a lot of unlock in there, like bringing my ideas together, right?

SPEAKER_01 28:29

Yeah. And it's a cool way of summarizing something and then using that output as context for something else, yeah.

SPEAKER_00 28:37

Yeah, it's it it's wild, and it's one of those things too. I'm um thinking about like, oh, what is what is gonna look like for education? Like if I had this set of tools available to me when I was in high school or college, one, I'm really glad I didn't because I feel like I would just never get anything done. Um and two, I don't think that the schools and universities have quite figured out how to give them the right power without letting them run the whole thing. Right. I think that like the best example I've seen is like we're gonna evaluate the AI's output. I'm like, that's the way to do it, but the rest of school is still around knowledge regurgitation.

SPEAKER_01 29:21

I actually wanted to ask you about customer education specifically because that's your wheelhouse. Um and we've had, you know, I think it's probably been a couple years since I saw the first neat trick of Hey Gen being able to translate videos on the fly. And you know, we've had these little things here and there that have helped

Customer Education In An AI World

SPEAKER_01 29:45

with you know accessibility and all of that. Um LMSs are a dime a dozen. And I I wonder what you think the future is for customer education, whether it's still this kind of siloed, you put these videos in the course and then you do a thing, and then or is it like we're building this stuff from scratch in a in a in in more of a web-based experience, or like what does that look like?

SPEAKER_00 30:18

I I think, and I'm gonna hedge a little bit here, uh-huh, because predicting the future is sort of messy, but I think it's gonna be a little bit of everything, and I think you're gonna see a whole spectrum of things where it's like when it needs to be 100% right, or there's a really high value, you'll probably still have a fairly like concentrated human producing or compliance, for instance. Compliance, it's like it's still gonna be pretty tight on the human rails, but I think you're also gonna see things where, and we're already seeing this, where the AI has been fed the content in a way it can understand and put together. And then you're gonna see maybe like you know, embedded chatbots where you can just chat about it, or almost like an ambient watching where it's like, it looks like you're struggling. Here's some just in time education. I think that's a really exciting way to think about it. But then I think the other piece, too, with customer education is having folks with expertise in customer education be the ones designing and working with the interfaces to make it so the learning is embedded within it. So UX design meets customer education. I think that's gonna be a lot of emerging. And I think there's still a lot of need for people who can explain this is how you teach a concept. Um, because it comes back to that same, you know, craftsperson idea of a really, really skilled instructional designer is going to get a better course out of course creation tools. Yeah. Um, and one thing I really like, uh, I've tried a lot with PARTA, which is you know, course creator uh tool. Um, for those who are unfamiliar, it's basically like Rise, but it's better. Um But we we talked a lot about their AI course creator tool. Um, and one thing I really like with the language in all these discussions is they landed on calling the output a draft. Cool. And it's like that is a very interesting thing, one because it just shows like respect for the craft. It's like it knows it's not going to be ready to publish. Yeah, it's not promising that, right? And I think that's where I'm I think that's where education is gonna be is these tools are gonna create drafts, and then the job of the customer education professional is to put in that last mile. Where, which surfaces do we put it on? What formats do we do? What do we have to record? Like, what do we have to make a human video on? How do I teach this live? I think that's the other frontier of customer education. I think as good as any asynchronous tool can be, uh, like I still sometimes fly out to a company and do like I'm in the same room with you. In-person training, we're going to look at the computer together for sure. And it's like that seems silly in some ways, but also there's such a value of it where we're in that same space, we're solving the same problem. That's where I see a lot of customer education dog.

SPEAKER_01 33:19

But it's like it's like superpowering the Khan Academy thing where you know we have all these tools where instructional designers can throw these things together very quickly and efficiently, and then but that then leaves more time for actual customer face time, you know. Um, and I'm I'm all I'm with you on the on-site. Like face-to-face time is so important, and that's not going anywhere.

SPEAKER_00 33:49

Yeah. I mean, the other major factor with all this is uh okay, we've got developers with all these tools are developing at an absurd rate. I mean, product changes, product features are coming out super fast. Yeah, the existing issue already for customer success, customer education is we can't keep up with it. Yeah, these tools are gonna let you keep up with it and build high quality content.

SPEAKER_01 34:13

Yeah, is there a good I mean, it you you don't have to name names. Okay, we'll talk about later, but if you don't want to, but like that specific problem of you have version whatever of your SaaS app, and then a week later you're like, you know, three micro versions later, and a bunch of stuff has changed. Like, is there something good off the shelf right now that tracks that and helps you create the content, whether it be knowledge-based articles or courses to or updating, you know, courseware, or is that still kind of I haven't seen anything off the shelf.

SPEAKER_00 34:53

What I've done before and what I've seen done, and it's actually pretty doable at most companies, is you look where your developers are pushing their commits and have an AI agent or something tracking that because then you are getting what is different, whether they've told you or not. Um, and also like where are the designers putting their PRDs? Like you have all that context, and with these AI tools, like you don't have to be actively watching it yourself. Like set up where it should be, set up like understand how engineering a product builds things, watch what the engineering managers are watching, and then apply your own filter to it. So, hey, every time you know a commit gets pushed through, check to see which surfaces it touches. And I also would map okay, which surfaces do my courses cover? Like my basics course covers this surface, this surface, and this surface. If anything changes there, depending on the level of it, I have to change it out. Yeah, and I think also just building your courses to be more resilient to that. Um, and this is again another principle. It's what's not going to change. Exactly. Uh, because it's like, okay, I know I'm gonna have a home page image. Yeah, like this is what the homepage looks like. I know I need to keep that up to date, yep. But I know there's always a consistent spot I can get it. So like that's the other thing is like be as lazy in a way as you can't anticipate like this is going to change, this is not going to change. If it changes, this is where I'm gonna take care of it and automate that.

SPEAKER_01 36:32

It's that age-old thing where you consciously have the voiceover omit the bit about the button being in the top right, because if you do that, I guarantee you the button's gonna move to the bottom right, you know. Oh, yeah, exactly.

SPEAKER_00 36:49

Well, and and I think the other thing we're seeing though is like tools like uh Cluso and Descript and all these video editors, even like a huge Clueso user. Yeah, they're so good, they're so good. They're like, yeah, also just the nicest. Yeah, but it's like you just see, oh, I don't have to worry about getting it exactly right because if it changes, yeah, it's not like I have to spend 10 hours in the like recording booth again and the edit it just to change like one thing. It's like, oh, we can change it. Yeah, and as even playing with a few like video production APIs over the weekend, it's like, oh, I can programmatically just tell Claude to tell me to change it, to tell my video editor to change it. Yeah, it's like that's just feels like the future. It's mind-boggling, yeah. Mind boggling.

SPEAKER_01 37:37

Well, um, we're gonna start rounding out the convo, although I think we could probably stay here for like freaking hours. I mean, I mean, yeah, every time you and I talk, it's just like, well, have you seen whatever? Yeah, we haven't even opened the laptops yet. Um I I would love this, might be putting you on the spot a little bit, but um there's like I I'm a fan of good better best. And for someone who's uh, you know, who's into wanting to make their own tools, you know, I I kind of have my own ecosystem of good, better, best or whatever, but I would love to hear from you, like, okay, you're a your basic, you're you're kicking the tires on vibe coding stuff and trying to get into it. Like, what does that look like? What is level two and what does level three

Good Better Best Vibe Coding Stack

SPEAKER_01 38:28

look like?

SPEAKER_00 38:29

Yeah, so I think the good is actually something we tried for the first time this weekend. Uh Google just released this new thing, I think it's called Sketch. Yeah, it's it's it's really good. It lets you prototype apps and websites, like web apps. It's like that's very, very well done. And their documentation is also really good. And what I like about that documentation in particular is it's opinionated. It tells you this is what we actually recommend and do, and here's other ways to do it. But like here's a golden path, which I think if you're just getting started out, like what a great way to do it. Yeah, and then they also have their AI studio, which is kind of that like next step up. Yeah, but there's a ton of like free credits with that, which I'm like, okay, if you're dipping your toe in and you like you could try it without really having any sort of commit.

SPEAKER_01 39:16

And like the Google ecosystem is so good for beginners, it's so good.

SPEAKER_00 39:21

And also like notebook on, like we talked about earlier, like use that because you can build pseudo apps with that kind of uh piece there. Um I would also be remiss to not say Zapier because I'm like and I'm not trying to sell Zapier, I just love Zapier. Totally. So it's like yeah, that that's there. Like there's so many cool things that I like, I spend most of my day just building with Zapier based. So I'm like, okay, I'm not trying to show, but I also love it. Yeah. Um, and then I think the kind of the next level up is even just using like those chat-based tools, like just using plain clot or plain chat to D or Plain Gemini to say, help me design this. Um, so like before they started building, it's like, help me code a website. I'm gonna have to put those pieces together and push them up to my GitHub. But like help me understand how to build these things and using that language. It's like, cool. Now I'm gonna get into a little bit more of the mechanics, right? Yeah, and then I think the kind of the premiere, and I can be seeing it, you know where this is going. It's it's clawed code, it's codex, it's cursor, it's that ilk. Antigravity, antigravity, yeah. Where you are taking that powerful brain of the AI tool of your choice. Yeah, you're giving it files in context. Um, another trick I'm doing this with cursors, MCPs. If you are not using MCPs, it they are the most slept-on thing out there. Uh again, Zapier, I'm biased on it, but it's like, oh, cool. Yeah, like that's 90% of my tools. I just connect with that. Um, but then like I connected the Chrome DevTools MCP. My cursor is able to browse the web, pull down websites, and do things for me. So it's like that whole set of tools together lets you output things that are unthinkable. Yeah, I had a project I was working on uh last week where I've had this idea in my head for probably two years now. I sat down, had a few hours just chatting with it, went, took a swim, came back, and was like, okay, I thought some more on it. Here's some more pieces, here's like the web framework I want to use. I was like, use Tailwind CSS because I wanted to look pretty. Here's the color scheme I want it in, and here's how I want it organized. It built this entire knowledge base and website and app around an idea I've had sitting in my head. Cool. And like that was impossible. Yeah. Like even in December. Yeah, yeah. Yeah, in December, it was impossible. And that's, I think, the other thing too for folks who are like, Oh, I you know, I tried those things on like my winter break. It's the latest generation of models, and I feel like I'm saying this next year, have changed the game. Yeah. Where like things that would have failed, that absolutely did fail, yeah, with the previous models, right? Now just get one shot. Like I was like, I want to build a mobile game that helps me explain concepts of Zapier. I'm like, here's the game design. Like I had that all written out, it was very well written. Every time I ran it through like the highest end models, I spent like 20 bucks of credits, which is a lot of credits. Um failed. Yeah. I put it through the new Chat GPT in February. Yeah. One shot. Really? Yep. How about that? And it was like, okay, that is really changing how I think about these tools. We have reached kind of that next level. Um, so that that I think is super exciting. So that's kind of the good, better, best is I like Google Google set of tools and then play with the models and have them tell you how to build stuff manually and then get in these frameworks.

SPEAKER_01 43:00

We're we're gonna close out on that one because like the the thing that people sleep on is it will tell you in the conversation, and in fact, by having the conversation about how should I do this, you are providing more context, and it's probably gonna lead to some follow-up questions and some clarifications that come out of that that you wouldn't have had otherwise. And if you're actively building stuff or prompting or doing whatever, like ask it to gut check itself, ask it to gut check you, because I guarantee you it's gonna come up with oh, we could probably do it this other way too, or this might be a little bit better. You know, the first shot isn't always the right thing to do. And a lot of people just blindly go through and go, Oh, yeah, yeah, yeah.

SPEAKER_00 43:55

Never never trust it and like never trust your own outputs, too. It's like you've got it, you've got this great tool. Uh there's like that whole Ted Lasso, like be curious, not judgmental. Like that is what you have to have in your brain when you're using AI. And uh, one last kind of tip for the like anyone who's listening, it's like it feels wild, but I almost never type to my AI tools anymore. I just use like you can use the built-in recorder tool, I use Whisperflow. But when you are talking to these AI tools directly, instead of trying to type it out, you're gonna get so much more context in. And they also just handle stream of consciousness. You you do not have to like prompt engineer anymore because by telling it exactly what you want, you're putting those elements of prompt engineering in there. It will parse it well enough.

SPEAKER_02 44:42

Yeah.

SPEAKER_00 44:43

And it's a game changer. It's gonna feel super weird for the first month. Like you will feel like crazy. Yeah. When my yeah, my wife walked into my office and she was like, Who are you talking to in that way? I was like, Oh, I'm talking to my AI. She's like, Oh, the robots again. Uh, but it's like it completely changes how these models work for me. Yeah, because it's like I'm having a conversation with a coworker. Yeah, I am not typing commands to a computer. Yeah. And that that switches the brain in a way that I think is powerful.

SPEAKER_01 45:22

I agree completely. Well, dude, it's always a uh an unlock for me when I chat with you, and I super appreciate it. And I think we should just have a recurring like invite on the calendar because selfishly I learn a lot when I talk to you. But um, yeah, it's awesome having you back. And maybe you'll be our first. I don't think there's any three timers. Ooh. Maybe you'll be our first three.

SPEAKER_00 45:46

Do you start getting like the robes like SNL?

SPEAKER_01 45:47

Yeah, we'll do a purple jacket for it. Perfect.

SPEAKER_00 45:49

There we go. Thank you for joining, Eric. This is great. I uh appreciate it. And like, hey, when in texts?

SPEAKER_01 45:57

Hey, in person.

SPEAKER_00 45:59

Yeah. Amazing. Thanks, I hope.

SPEAKER_01 46:01

Thanks. Thank you for joining me for this episode of the Digital CX Podcast. If you like what we're doing, uh consider leaving us a review on your podcast platform of choice. If you're watching on YouTube, leave a comment down below. It really helps us to grow and provide value to a broader audience and get more information about the show and some of the other things that we're doing at digitalcustomersuccess.com. I'm Alex Drogovic. Thanks so much for listening. We'll

Final Takeaways And How To Support

SPEAKER_01 46:26

talk to you next week.