Latest / The Same Side: A Digital Marketing Podcast Rooted in Human Connection / Episode 11: AI Is Only as Good as Its Context: 2026 MAICON Recap

Transcript

Episode 11: AI Is Only as Good as Its Context: 2026 MAICON Recap · The Same Side: A Digital Marketing Podcast Rooted in Human Connection

• 0:10
  SPEAKER_01: Hey there.
• 0:11
  SPEAKER_01: Welcome to the same side, a digital marketing podcast rooted in human connection.
• 0:15
  SPEAKER_01: I'm your host, Jonathan Dance, Director of AdOps and Analytics at Digital Relativity, a full service marketing agency.
• 0:22
  SPEAKER_01: You can find us on the web at digitalrelativity.com.
• 0:26
  SPEAKER_01: A few of us just got back from MAKON, the marketing AI conference in Cleveland.
• 0:31
  SPEAKER_01: And after we got back, we talked through what we learned in our all-team meeting and we used an AI tool to take notes.
• 0:37
  SPEAKER_01: And as with all great AI output, it was weird because it called the conference MCON the whole way through.
• 0:44
  SPEAKER_01: And that's relevant because one of the things we kept coming back to at the conference is that AI is only as good as the people using it and the people who understand the difference between a good output and a not so good output, the people who know the context essentially.
• 0:59
  SPEAKER_01: So that's what today's episode is about.
• 1:01
  SPEAKER_01: And I'm joined by the two people that I went with: Aaron Gooden, full stack developer and ad ops strategist, and Pat Strader, our founder and chief innovation officer.
• 1:11
  SPEAKER_01: Welcome, guys.
• 1:12
  SPEAKER_01: Hey Jonathan.
• 1:13
  SPEAKER_01: Hey, good to see you guys again.
• 1:16
  SPEAKER_01: Yeah, do you guys want to just give a little bit of intro about yourselves and maybe just how you use AI on the regular?
• 1:23
  SPEAKER_01: All right, Pat, you go first.
• 1:25
  SPEAKER_03: Yeah, I'll jump right in.
• 1:26
  SPEAKER_03: AI, you you can't go anywhere now without there being conversations about AI and the differences and where people are with how they're using it and the understanding of it.
• 1:38
  SPEAKER_03: Is it's a really interesting time.
• 1:40
  SPEAKER_03: And it's something that I've been paying attention to and how it's going to impact the work that we do as an agency, but also the work that our partners do.
• 1:54
  SPEAKER_03: And so this has been something that I think that we've all been following pretty closely here as an organization at Digital Relativity, trying to keep our partners informed as to best practices and ways to use these tools.
• 2:09
  SPEAKER_03: And so it's just a it's a really exciting time, and not trying to get ahead of ourselves here, but this conference, MAKON, I think this is my fourth that I've attended, and it was just really interesting to see the conversations and how they've evolved related to how we use AI.
• 2:26
  SPEAKER_01: Yeah, cool.
• 2:27
  SPEAKER_01: I'll look forward to getting into that here in a bit.
• 2:29
  SPEAKER_01: Uh how about you, Aaron?
• 2:31
  SPEAKER_02: As you said, I'm a full stack web developer here at Digital Relativity, and I've been interested in AI for almost my whole life, and not saying that as if I created it or thought of it before anybody else, but it's been the dream of computer scientists to have this thing where the computer becomes almost human.
• 2:54
  SPEAKER_02: And I can remember back in the early 80s working on Commodore 64, Commodore Big 20s, and dreaming of a day when you would be able to talk to your machine and it could do things for you.
• 3:06
  SPEAKER_02: The issue is that when I was in elementary school, I really didn't think through too many of the consequences, the ramifications of having that kind of technology.
• 3:15
  SPEAKER_02: And so what we have today is the outgrowth of that idea that's come to pass in our world, and we're having to deal with how to use it responsibly and how to use it uh within our organization.
• 3:28
  SPEAKER_02: And so that's something that I'm working on and interested in.
• 3:31
  SPEAKER_02: Awesome.
• 3:32
  SPEAKER_02: Yeah, MACON was a piece of that.
• 3:35
  SPEAKER_02: So well, cool.
• 3:36
  SPEAKER_01: And so a little context about MAKON.
• 3:39
  SPEAKER_01: Speaking of context, it's a three-day conference about AI and marketing, and it's put on by the AI Marketing Institute, headed up by Paul Ratzer and Mike Kaput.
• 3:48
  SPEAKER_01: We saw speakers that range from people from Google and big agencies to journalists and authors, and the sessions that we had at our disposal included hands-on workshops, big picture keynotes, and about where we think this is all heading, right?
• 4:04
  SPEAKER_01: We and we'll get into this again a little bit more in our takeaways, but just to Pat's point, how things have changed over the years going to this conference.
• 4:14
  SPEAKER_01: I think one of the biggest things that struck us, and probably you guys more, this is only my second make on, is that the conversation has moved past the hype and trying to figure out what tools to use and get our feet under us as an industry.
• 4:25
  SPEAKER_01: Two years ago, sessions were about awareness.
• 4:27
  SPEAKER_01: Here's what it is, here are the tools, don't get left behind.
• 4:30
  SPEAKER_01: Oh my gosh, you're getting left behind.
• 4:31
  SPEAKER_01: And this year the focus was more on refining how we use those tools as the tool landscape has settled down significantly and how we use those in our actual day-to-day work.
• 4:42
  SPEAKER_01: With that in mind, Aaron, what was your big takeaway after this?
• 4:45
  SPEAKER_01: Is what your fourth, third or fourth?
• 4:48
  SPEAKER_02: This was my third.
• 4:49
  SPEAKER_02: Um, and I did see the progression through the hype and legalities and into the process of getting this into your organization and how to structure AI policy and so on.
• 5:04
  SPEAKER_02: And then this year was more refinement and how to use this technology well.
• 5:09
  SPEAKER_02: So the main takeaway that I had was really around this idea of context.
• 5:16
  SPEAKER_02: Context being the background that you give any AI system before you ask it to do anything.
• 5:22
  SPEAKER_02: The the example or the way we could talk about this is to say if you had a new hire in your organization and you asked that person to create something, it could be a visitor's guide or a brand guide.
• 5:33
  SPEAKER_02: You wouldn't just say create this, you would give them context.
• 5:37
  SPEAKER_02: You would give them examples of what you've done before.
• 5:40
  SPEAKER_02: You would hand them maybe last year's brand guide or transcripts of meetings that have something to do with what they're generating.
• 5:49
  SPEAKER_02: And so just to give them a context for exactly what you're looking for, and AI is no different.
• 5:56
  SPEAKER_02: You can think of AI as a new hire.
• 5:59
  SPEAKER_02: AI is very capable, but its knowledge is general.
• 6:02
  SPEAKER_02: It has a broad knowledge, but it's very general.
• 6:05
  SPEAKER_02: The more context you can give an AI, an AI agent of any kind in a conversation, the better off you're going to be.
• 6:13
  SPEAKER_02: That includes a lot of different things.
• 6:15
  SPEAKER_02: It can include conversations, various files, CRM data, and documented processes.
• 6:22
  SPEAKER_02: It could also include transcripts and so on that you can give the AI that so it could produce the the output that you're looking for.
• 6:32
  SPEAKER_02: This can be translated really into we talk a lot about claw because we use it, and it uses the word skill, and that's a word that's used across AI models, but it's just a set of instructions that stayed inside the tool that you can reuse, and the context can live there, or the skill can point the AI to the context.
• 6:55
  SPEAKER_02: At MACON, we've heard a lot about the different ways that companies are designing these context sources for AI, and skills is one of those.
• 7:04
  SPEAKER_02: And so I came away with a newfound interest in putting some skills together to make that happen.
• 7:09
  SPEAKER_02: So to create good context for us.
• 7:12
  SPEAKER_03: Well, and the context is critical.
• 7:14
  SPEAKER_03: And that that's what helps provide the better outputs.
• 7:18
  SPEAKER_03: Taking a few steps back here, the first MAKON that I attended, I recall Mike put doing a session that was like, I don't recall exactly how many it was, but it was like 55 tools in 30 minutes or whatever.
• 7:31
  SPEAKER_03: And everybody's sitting there writing everything down as fast as they can, taking screenshots of what was on the screen.
• 7:36
  SPEAKER_03: And what we had was all these tools, but we would go to the tools.
• 7:40
  SPEAKER_03: There was no, first of all, there was really no instruction manual.
• 7:43
  SPEAKER_03: So it was like we'd just kind of type something in, do this thing.
• 7:47
  SPEAKER_03: Without any of the additional information, like brand guides and reports and the meeting notes and all the things you just described here.
• 7:56
  SPEAKER_03: Evolution, is that the right word?
• 7:58
  SPEAKER_03: The evolution of how these tools and things and how we use them has come about.
• 8:03
  SPEAKER_03: And so now the key is to how do we implement the context in the best possible way so that we get the best possible outputs that don't follow in the vein of the work slop, the slop grenades, all the things that we hear people talking about.
• 8:21
  SPEAKER_03: Jonathan, I think I heard you call one a slop rocket.
• 8:24
  SPEAKER_03: So how do we prevent those types of things?
• 8:27
  SPEAKER_03: It's the context.
• 8:28
  SPEAKER_03: And so now the things that we need to figure out is where do we put that context so that it's accessible to the tools and relevant to the tools too?
• 8:37
  SPEAKER_03: So if you just have one folder that has everything in there, you're making it work even harder.
• 8:42
  SPEAKER_03: And so that's where I feel this is where I'm going to be going now is how do we architect this context and map it to the tools and to the people using the tools in the appropriate context for their job, their role, their project, whatever the case may be.
• 9:01
  SPEAKER_03: So there's it's it's exciting to see this kind of move from just find a tool, go type something in and see what you get.
• 9:10
  SPEAKER_03: And it's just really it's great to see the refinement in the outputs.
• 9:14
  SPEAKER_01: Yeah, for sure.
• 9:15
  SPEAKER_01: You know, I remember back in 2023, we went to PubCon, and that was right at the moment, like the inflection point, at least in my world of AI actually being a usable tool versus being and and granted, it was crude then, but the one thing that's interesting is the the evolution too, is oh, we you're gonna need to be a prompt engineer, right?
• 9:35
  SPEAKER_01: Like a prompt, and it's like we're beyond that now because it's like, okay, you still need to have the prompts, but now you have these skills, or what do they call them in ChatGPT now?
• 9:44
  SPEAKER_01: Anyway, it doesn't matter.
• 9:46
  SPEAKER_01: It'll probably change tomorrow.
• 9:48
  SPEAKER_01: So now it's kind of you can build it once and see the output and then refine it.
• 9:52
  SPEAKER_01: So you do all your prompting up front for these tasks that you need to do repeatedly, and it's just gotten so much better, but that the context is still important.
• 10:01
  SPEAKER_01: And then my takeaway is that in order to judge your outputs, kind of going back to the beginning of the intro, is that you can't just take it, right?
• 10:09
  SPEAKER_01: Oh, well, hey, I got this output and now I'm going to give it to our partner because inevitably AI is still artificial and it's not really intelligence, right?
• 10:17
  SPEAKER_01: It's a language model and it learns based on what already exists.
• 10:21
  SPEAKER_01: And what we've also learned, I think, is that it skims.
• 10:24
  SPEAKER_01: It's not like it's, hey, here's all my stuff.
• 10:27
  SPEAKER_01: Tell me what the observations or recommendations can be from all this stuff.
• 10:31
  SPEAKER_01: And it's like, yeah, here's here you go.
• 10:33
  SPEAKER_01: And it's like, what about this?
• 10:35
  SPEAKER_03: Yeah.
• 10:36
  SPEAKER_03: I always like it when it's like, that's going to take me a really long time.
• 10:39
  SPEAKER_03: And it's like, what?
• 10:41
  SPEAKER_01: Do you do you feel like it's like I really don't want to do it because I'm I'm lazy, but if you want me to, I will.
• 10:47
  SPEAKER_01: And so I think that's the most critical part.
• 10:50
  SPEAKER_01: So the context is key because that's coming from us as well, right?
• 10:53
  SPEAKER_01: The the ability to shape its responses based on our specific needs.
• 10:58
  SPEAKER_01: And then also when it comes out, also using our expertise and our historical knowledge of our partners and all that stuff to make sure that what's coming out is relevant to that.
• 11:10
  SPEAKER_01: Just a quick aside, I think, of a lot of campaign reports that would do, and we'll run it through Claude, and it'll be like, oh man, you really killed it in May, but something's different now.
• 11:20
  SPEAKER_01: And it's, yeah, we weren't running paid media this month, but in May we were running a ton of paid media.
• 11:24
  SPEAKER_01: And so you can't just kick back and let AI do your work for you.
• 11:27
  SPEAKER_01: You still have to apply what you know.
• 11:29
  SPEAKER_01: I don't need to get any deeper into that takeaway, but you got to bring your experience to the table when you're working with AI.
• 11:35
  SPEAKER_01: And the other thing I think is also important in our particular realm in the marketing realm is that you need to understand your partner's goals too.
• 11:43
  SPEAKER_01: And this goes back to your guys' context stuff, and ideally, you have that baked into your libraries.
• 11:48
  SPEAKER_01: All those takeaways are interrelated, right?
• 11:51
  SPEAKER_01: They interdigitate.
• 11:53
  SPEAKER_01: So, how do you manage this stuff?
• 11:55
  SPEAKER_01: Do you do you have a central hub like we're talking about?
• 11:58
  SPEAKER_01: Who manages it?
• 11:59
  SPEAKER_01: Pat, I don't know if you want to speak to that a little bit.
• 12:01
  SPEAKER_03: That's the challenge for organizations of all different types.
• 12:05
  SPEAKER_03: Where does that responsibility lie?
• 12:08
  SPEAKER_03: First of all.
• 12:09
  SPEAKER_03: But then secondly, where does it live?
• 12:12
  SPEAKER_03: How is it mapped to the tools, to the people, as we mentioned a moment ago?
• 12:19
  SPEAKER_03: And then who maintains it, right?
• 12:21
  SPEAKER_03: And at MakeCon, we kept hearing the talk about the builder, right?
• 12:27
  SPEAKER_03: The organizations will have the builder that's replaced the prompt engineer.
• 12:32
  SPEAKER_03: I'm using air quotes here.
• 12:34
  SPEAKER_03: The builder is the one that helps make the tools, the internal tools, okay.
• 12:40
  SPEAKER_03: And in my mind, it requires a builder, but it absolutely, I'm gonna channel my inner Justin here, and it's like building a house, right?
• 12:50
  SPEAKER_03: So you don't just hire a construction company and say, build me a house, right?
• 12:55
  SPEAKER_03: You have to give them the plans.
• 12:57
  SPEAKER_03: Not necessarily an architect, but I'm gonna use that anyway.
• 13:00
  SPEAKER_03: You have to give them the structure and the plan of here's how this thing is to be built.
• 13:06
  SPEAKER_03: So that's the next challenge I I see for organizations, again, like us and as an agency, but also many of our partners, is where does the responsibility lie?
• 13:18
  SPEAKER_03: How often is that that context updated?
• 13:23
  SPEAKER_03: How is it updated?
• 13:24
  SPEAKER_03: Who's doing the updating?
• 13:26
  SPEAKER_03: There's just all these different questions that need to be answered.
• 13:29
  SPEAKER_03: And we like to call ourselves problem solvers, and that's why I find this this particular time really exciting.
• 13:35
  SPEAKER_03: And just as a quick aside, talking about how this has changed over the years, you made the comment about PubCon.
• 13:43
  SPEAKER_03: I'd actually had forgotten about that.
• 13:44
  SPEAKER_03: PubCon has a conference that I've been going to for more than 20 years, and and it's like a gathering of some of the best and brightest digital marketers on the planet, honestly.
• 13:56
  SPEAKER_03: And at that conference, what year did you say that was?
• 13:58
  SPEAKER_03: Was that 23?
• 13:59
  SPEAKER_03: 23.
• 14:00
  SPEAKER_03: Okay, 23.
• 14:02
  SPEAKER_03: And here are these incredibly bright and smart people.
• 14:05
  SPEAKER_03: I'm not picking on anybody at all.
• 14:06
  SPEAKER_03: And in fact, I give a ton of credit to everybody there because we were discussing it early on that people were referring to it as chat GTP.
• 14:14
  SPEAKER_03: That's just how quickly things have changed.
• 14:17
  SPEAKER_03: How we went from what do we even call this thing chat GTP to what do we do with it?
• 14:24
  SPEAKER_03: So it's an exciting time to be alive.
• 14:27
  SPEAKER_01: One thing that also strikes me is we've also gone from, oh, I've got this tool, I'm just gonna monkey around with it and see what I can get it to do, to we need an organizational framework for this.
• 14:37
  SPEAKER_01: And so as with any any good fun thing, we're gonna kill the joy out of it by turning it into a bureaucracy, but it's necessary, right?
• 14:45
  SPEAKER_01: To make sure that it's firing the way it needs to and serving us the way it needs to.
• 14:49
  SPEAKER_01: That's right.
• 14:50
  SPEAKER_02: I think we've moved away from what you guys are talking about in the quote early days, two years ago or three years ago, whatever it's been.
• 14:57
  SPEAKER_02: It used to be that if you used AI, that was like your differentiator.
• 15:01
  SPEAKER_02: You you actually used it, or if you used the best model that was out at that time, that was your differentiator and made you stand out.
• 15:10
  SPEAKER_02: That's no longer the case.
• 15:11
  SPEAKER_02: Everybody has access to the same models, the same setup, and so really the context is the differentiator.
• 15:18
  SPEAKER_02: And I would add to Pat's illustration of the house.
• 15:22
  SPEAKER_02: When you have a house built, you have an architect design it and put the plans together, and then you have somebody build the house.
• 15:29
  SPEAKER_02: But equally important is this maintenance idea, and you really need to hire the people that mow your grass and keep it painted and everything clean and maintained.
• 15:38
  SPEAKER_02: We've heard this idea of a slop grenade mentioned in one of the presentations, and we really need to think about that more as a slop rusting or a slop reverb, where there's this feedback mechanism that can happen in your context if you allow the model to create things and then those aren't maintained or maybe even deleted sometimes, so that the context is always something that's human-driven as opposed to AI created, so that it doesn't feed back on itself and create slop rust in your content.
• 16:16
  SPEAKER_01: Would it be like the Borg?
• 16:17
  SPEAKER_01: So, like eventually it's just gonna assimilate everything.
• 16:21
  SPEAKER_01: Yeah, I think that makes a ton of sense.
• 16:24
  SPEAKER_01: As things continue to evolve, we're trying to stay ahead of how to manage all of it, and that's gonna continue for sure.
• 16:30
  SPEAKER_01: And in order to do that, obviously we've come back from MACON and we have ideas, right?
• 16:36
  SPEAKER_01: We went to some pretty good talks.
• 16:38
  SPEAKER_01: And so I guess for closing this episode, what are you guys doing to take some of the things that we've learned at MAKON, talking about the context and that sort of thing?
• 16:46
  SPEAKER_01: Uh, you know, in 60 days, where do you see yourself and those tools or the organization being?
• 16:52
  SPEAKER_03: Thinking about the context and where does it live, right?
• 16:55
  SPEAKER_03: And how do we help members of our team, our partners, the people that we work with use these tools and understand where that goes?
• 17:03
  SPEAKER_03: And we haven't solved that yet, but we will.
• 17:06
  SPEAKER_03: So it's trying to determine what is the structure going to be.
• 17:11
  SPEAKER_03: Where is that context going to live?
• 17:13
  SPEAKER_03: For example, right now, we can go in, you and I individually can put in our own custom instructions, if you will, that gives the context of this is who I am, this is my role, don't use these words, those kinds of things, don't put delve into anything, that type of stuff.
• 17:32
  SPEAKER_03: But then as people or teams, I should say, individuals as well, are becoming more sophisticated in how to use these tools, there's more options for where that context can live.
• 17:43
  SPEAKER_03: So I'm going to be working on developing the framework of where we internally can put that context, but then also to help our partners, right?
• 17:54
  SPEAKER_03: To help our partners understand how they can set up their tools for their work.
• 18:01
  SPEAKER_03: Awesome.
• 18:01
  SPEAKER_01: Aaron, how about you?
• 18:02
  SPEAKER_01: What's cooking in the Aaron kitchen?
• 18:04
  SPEAKER_02: A couple of things.
• 18:05
  SPEAKER_02: The first thing is in my department, which is the development and ad ops department, we are beginning to look at very closely the context content that we have and trying to figure out what putting that together in a way that can serve as general context for our work specifically, what that is going to look like.
• 18:26
  SPEAKER_02: And so we're working on that.
• 18:28
  SPEAKER_02: And then also I've been developing some tools for us to use with task management across the company.
• 18:34
  SPEAKER_02: And that started as a personal work for myself to keep myself straight with tasks.
• 18:40
  SPEAKER_02: But more and more I'm understanding that the context, again, is really important for that work, trying to figure out how to better feed the AI with the data from multiple streams in order to get the right results.
• 18:53
  SPEAKER_01: I remember a conversation that we had, maybe it was even on the way home, but the idea of you have our organizational context, so DR, what are we about?
• 19:02
  SPEAKER_01: What are our goals?
• 19:03
  SPEAKER_01: How are we trying to serve our partners?
• 19:04
  SPEAKER_01: And then there's the individual partner profiles, for lack of a better term, that need to be layered in there too, right?
• 19:11
  SPEAKER_01: I was sitting in one of the keynotes, and I confess I wasn't 100% tuned into the keynote, but I was building something.
• 19:18
  SPEAKER_01: It started out as a media plan builder, something to generate, something that we could work off of an adops and obviously fine-tune, add our own experience and context to.
• 19:29
  SPEAKER_01: But the thing that struck me when I was doing it, and I was tapping into a lot of different sources, tapping into BigQuery, but the power of Claude to be able to take historical data and provide benchmarks for that partner, for individual ad platforms that we could we've always used industry benchmarks as a gauge of success, but now being able to know what success looks like for that individual partner without having to spend hours upon hours of crunching numbers and hoping that it's right.
• 20:00
  SPEAKER_01: Because the nice thing is once you get these outputs, at least from the ad op side, we can check them.
• 20:04
  SPEAKER_01: I can go into GA4 and see, do a little bit of quick math to spot check things and go, okay, yeah, that that actually looks good.
• 20:11
  SPEAKER_01: And then the other tool that I'm currently working on is one for me for managing purposes, is to I won't go too far in the weeds in this, but we have a couple tools in adopts that we use.
• 20:21
  SPEAKER_01: One is a giant sheet that has all the campaigns and pacing and all this stuff.
• 20:25
  SPEAKER_01: And it's no one likes to look at it.
• 20:27
  SPEAKER_01: It looks like a slop grenade, but it's super useful.
• 20:30
  SPEAKER_01: But then the other one, we just started implementing a daily check-in sheet that has basically every project that everyone in the group touches.
• 20:40
  SPEAKER_01: And we stole it from the creative department and our slop grenade.
• 20:45
  SPEAKER_01: It is a slop grenade.
• 20:46
  SPEAKER_01: We touch a lot of projects.
• 20:47
  SPEAKER_01: And the problem is that our most of our projects are not discrete one-offs, like, oh, hey, make this thing.
• 20:53
  SPEAKER_01: Okay, I'm done.
• 20:54
  SPEAKER_01: Click and it's done.
• 20:55
  SPEAKER_01: It's like, hey, here's a campaign that's going to run for 12 months.
• 20:57
  SPEAKER_01: And so you it just builds a bills.
• 20:59
  SPEAKER_01: So anyway, trying to figure out a way to automate some of that, but also make sure that stuff isn't falling through the cracks, right?
• 21:06
  SPEAKER_01: That's always the challenge when you're touching a lot of things, is that sometimes things can fall by the wayside.
• 21:10
  SPEAKER_01: So in 60 days, my hope is to have something that's usable that helps me in our daily meetings to make sure we're on top of everything.
• 21:18
  SPEAKER_01: So super mundane, but I think could be a huge game changer in terms of efficiencies.
• 21:23
  SPEAKER_01: And that's kind of it.
• 21:24
  SPEAKER_01: So any last parting words, one sentence from each of you guys, or whatever, you know.
• 21:30
  SPEAKER_03: Don't be a slop slinger.
• 21:33
  SPEAKER_03: How's that?
• 21:34
  SPEAKER_01: I like it.
• 21:34
  SPEAKER_02: Don't become part of the problem, become part of the solution.
• 21:38
  SPEAKER_02: I I would say if you haven't built a skill, go build one.
• 21:42
  SPEAKER_02: You may need to hook into some wisdom from the people that are building some of this stuff for your organizations.
• 21:49
  SPEAKER_02: But the skill of building a skill is important and one to have.
• 21:52
  SPEAKER_01: So go do it.
• 21:54
  SPEAKER_01: I'm glad you said that.
• 21:55
  SPEAKER_01: One of the things that I think about when building skills, and that I'm usually it usually takes me a second to go, oh, right, right.
• 22:01
  SPEAKER_01: You can have Claude help you build a skill, and of course, you have to provide the context, but you don't have to come up with it out of whole cloth from your head, which is a nice thing.
• 22:12
  SPEAKER_01: The last thing I'll say is that I'll be curious to see where we are 12 months from now and how we're talking about this stuff and what kind of changes we will have seen.
• 22:21
  SPEAKER_01: That's the nice thing about going to MAKON is that it's a conference, and just like most conferences, you got to separate the wheat from the chaff or whatever.
• 22:29
  SPEAKER_01: But every time, all two times that I've gone, I've come away not just with big ideas, but also very translatable things that I can take into the workplace and implement that aren't pie in the sky.
• 22:42
  SPEAKER_01: They're just like, yeah, you just get in there and get to it.
• 22:45
  SPEAKER_01: And then because all three of us have gone and because we have a robust usership of Claude and other AI tools here at DR, we have a nice context for developing this stuff and gut checking things.
• 22:57
  SPEAKER_01: And so hopefully, after listening to this episode, maybe you've got some ideas that are cooking in your head.
• 23:03
  SPEAKER_01: If you haven't been to MACON, highly recommend going.
• 23:06
  SPEAKER_01: That's it.
• 23:07
  SPEAKER_01: That's the episode in a nutshell.
• 23:09
  SPEAKER_01: Thank you so much for listening.
• 23:10
  SPEAKER_01: Aaron and Pat, thank you guys for coming on.
• 23:12
  SPEAKER_01: We'll have to do this again sometime.
• 23:14
  SPEAKER_01: This was fun, hopefully sooner rather than later.
• 23:16
  SPEAKER_01: Thanks for letting us hang out.
• 23:18
  SPEAKER_01: Absolutely.
• 23:19
  SPEAKER_01: All right.
• 23:20
  SPEAKER_01: Again, thanks for listening, and uh, we'll see you next time on the same side.