--- title: "AI Can Do the Work, But Should It? with Adam Larsen" id: "15430" type: "podcast" slug: "ai-can-do-the-work-but-should-it-with-adam-larsen" published_at: "2026-08-27T10:30:00+00:00" modified_at: "2026-08-27T10:40:49+00:00" url: "https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/ai-can-do-the-work-but-should-it-with-adam-larsen/" markdown_url: "https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/ai-can-do-the-work-but-should-it-with-adam-larsen.md" taxonomy_shows: - "Screaming in the Cloud" --- About the Author Corey is the Chief Cloud Economist at Duckbill, where he specializes in helping companies improve their AWS bills by making them smaller and less horrifying. He also hosts the "Screaming in the Cloud" and "AWS Morning Brief" podcasts; and curates "Last Week in AWS," a weekly newsletter summarizing the latest in AWS news, blogs, and tools, sprinkled with snark and thoughtful analysis in roughly equal measure. [https://podcasts.apple.com/us/podcast/screaming-in-the-cloud/id1361244178](https://podcasts.apple.com/us/podcast/screaming-in-the-cloud/id1361244178) [https://overcast.fm/itunes1361244178/screaming-in-the-cloud](https://overcast.fm/itunes1361244178/screaming-in-the-cloud) [https://pca.st/7l2e](https://pca.st/7l2e) [https://open.spotify.com/show/3fBA9eNkGliCzp3Xuy1GVd](https://open.spotify.com/show/3fBA9eNkGliCzp3Xuy1GVd) [https://feeds.transistor.fm/screaming-in-the-cloud](https://feeds.transistor.fm/screaming-in-the-cloud) ## Episode Summary ## Episode Show Notes & Transcript On this episode of *Screaming in the Cloud*, Corey Quinn sits down with Adam Larsen, Principal Developer at Aurora Educational Technologies, to explore what technology and AI should actually look like in K-12 education. They discuss using tech to empower teachers rather than replace them, the risks of outsourcing critical thinking to AI, student data privacy, and what happens when schools become dependent on tools that may not be affordable forever. Adam also shares how AWS became his go-to platform for solving complex education data problems, and why sometimes the best technology is simply the one that reliably gets the job done. **Show Highlights:** 00:00 Critical Thinking Warning 00:15 Meet the Guest 01:07 From Psych to Edtech 02:07 Teachers and Tool Whiplash 07:29 AI as a Teaching Aid 10:07 Process Over Shortcuts 14:29 Tech Misconceptions in Schools 16:38 Privacy and Picking Vendors 21:50 AI Costs and Ecosystem Bets 26:33 Building on AWS and Wrap Up **Sponsored by:** [duckbillhq.com](http://duckbillhq.com/) ### Transcript Adam: If we offload that critical thinking to a bot that then produces the outcome, we are doing ourselves a gigantic disservice Corey: Welcome to Screaming in the Cloud. I'm Corey Quinn. I'm joined today by Adam Larsen, who's a principal developer at Aurora Educational Technologies. Adam, thank you for joining me. Adam: Thanks for having me here. It's great to Corey: meet you, Corey. This episode is sponsored in part by my day job, Duckbill. Do you have a horrifying AWS bill? That can mean a lot of things, predicting what it's going to be, determining what it should be, negotiating your next long-term contract with AWS, or just figuring out why it increasingly resembles a phone number, but nobody seems to quite know why that is. To learn more, visit duckbillhq.com. Remember, you can't duck the Duckbill bill, which my CEO reliably informs me is absolutely not our slogan. You're doing one of those fun things where you work in edtech in the K through 12 space, but for some reason, you don't seem to view the students, which most of us call, you know, children, as commercial cattle to be sold things to. How'd you get there? Adam: I got there by not being a technologist at heart. So I came into education actually as a school psychologist. So I have a bachelor's and a master's in psychology, and I got into K12 as a psychologist and found myself solving very interesting problems that school districts hadn't really encountered before. So a lot of, um, pulling data from here, pushing data to here, making it so the teachers' jobs were easier. So students are data points only in so much as we like to try to provide information to people to make the best decisions they can about kids, and I found myself solving that problem over and over and over again. And then over time, I got more into data than I was into working with students individually, but never lost that focus on making it easier for teachers to do their job. Corey: I've been fairly involved in the, in our local school, US, uh, NSFUSD. Uh, my daughters are in the elementary school, and what I found interesting is even in the same school, some years the teachers seem to absolutely embrace technology, and others shy away from it. Uh, probably the most extreme example of that is, "Oh, it's time for parent-teacher conferences. Fill out a form of availability during this week and send them back in." The teacher will then build a chart and then send out home a note the following week with what days w- your appointment is. It's- I hear you, but have you heard of this thing called a Google Doc? I didn't even start on the Calendly approach or anything like that. And the answer was very, it was very polite, "I've been doing this for 30 years. Don't tell me how to do th- my job." Like, fair enough. Fair enough. Just this would be easier for everyone involved, but okay. And then you have folks who just absolutely embrace it to the point where I've talked to a multiple, a multiplicity of teachers who have been very annoyed by Google killing their Jamboard, uh, whiteboard product. And Google, the Googlers I talked to didn't seem to understand that. Like, we're giving them at least a year's notice. What's the concern here? It's, the concern is that they have built a lesson plan around this entire technology that you have, uh, been selling them for a while, and now regardless of how long that runway is, they have to redo something that was working for them. I, I don't think you appreciate just how much you have annoyed people with this approach. These are not technologists who are gonna find, oh, we've completely re-skinned your IDE. Time to learn. This is a cohort of folks who very often, if you change the color of an icon, they're confused and open a ticket. Like, this is not the core function of what they do, nor should it be. Uh, their, their only crime in this case was apparently trusting Google. Adam: Which as we've found out over the years, is not necessarily a great place to put your trust if you love little tools here and there. You know, there's a whole- Twitter account of Killed by Google, right? Just over and over again. Like- Corey: Cody Ogden lives here, does it. Uh, we had a, he and I, they, they and I have coffee periodically, and they are just, they're spectacular. But it's, it's the common problem every single time, where I, I used to evangelize Google products to my friends and family. These days, no, because I have to walk it back two years later. Adam: That's right. So a couple things there. Uh, one is that that lean in and then get burned cycle happens not just with technology, with teachers. They will get super excited about a new curriculum initiative or a new pedagogical practice, and they'll get super jazzed about it. And then another book comes out, or a new principal comes in, and top-down initiatives happen, and then they get burned. So what they thought was going to last them for a long time is suddenly the stinky potatoes of yesteryear, and then everyone looks at them like, "Why are you still doing that thing?" Like, well, five years ago, this was amazing and cutting edge. And then when you introduce the thing of like they're not even in control of it, like for example, Google kills something or there's some, you know, there's startups all the time that somebody gets excited about, the cash dries up, and they disappear, and they invested all of this time and energy. People, and teachers are, are no exception to this, they only have so much creativity and energy for creating, you know, new things. And I think that's- Teachers really like to create something and then be able to rely on it working again a year later. So if they build something and build several somethings and they're used to being like, "Okay, th- well, this unit I'm gonna do in this project," and that thing goes away, now, like I have to, like you said, I have to go back and rebuild that whole thing again, and I'm also going to be a little less likely to jump in and use the next hot thing when it comes along. Um, so that can be fairly scary for them. We used to say with tech that like, I don't know, maybe like 10, 15 years ago, like you even mentioned it when you were asking the question about like the age of the teacher. So s- we used to say like, "Well, you know, old teachers are stuck in their ways. They've done this for 30 years. I don't want to." But I've seen teachers right before retirement just as excited to embrace new things in their last two, three, four years. Corey: It's not an age thing. I think people lose sight of that. Adam: It absolutely isn't. Corey: But it also depends on how, uh, how teachers view what they do on some level. It's, it's Well, let's take podcasting, something that I know a little bit about. I, I ... There are some folks I talk to that are constantly chasing the next studio piece of equipment and the rest. Uh, this is our office. I, I'm currently using the next to last generation of stuff I have. First generation's in my studio at home, but I bought all this stuff during the pandemic. It, it's five years old. It works just fine. At some point, stop focusing on the tools you're using to create the thing, and just go ahead and have the conversation. D- uh, have the content be good rather than continuing to chase the, uh, ever, uh, ever, uh, removed point of diminishing returns. Adam: I'm reminded of a developer who fiddles with key bindings for days and weeks on end before they actually write any code. Right? That's- Corey: Oh, G- I, I tried to go down the custom keyboard well a while back. The problem is is I travel too much, and I take my laptop with me, that I constantly have to use the standard layout, and I don't have the mental wherewithal to pick up both. Adam: No, not to change gears like that, no. I, I would rather do, do the work rather than mess with my tooling for hours and days and weeks. Corey: Yeah, it's, it, it's one of those weird things. Uh, I have to ask, in this era, how are you approaching AI, especially in the K to 12, K through 12 space? I, it's evolving so rapidly. My kids are still young enough they don't really have to have those conversations yet because they're closer to K than 12, but that's changing. Adam: Yeah, so I have, uh, I have one squarely in the middle of that, a fourth grader. Um, and we had back to school night last night and found out that, uh, because he's up in the fifth grade for some of his subjects, he actually gets to take his Chromebook home for the first time. Uh, they're right on that cusp, and since he's kind of straddling it, he gets to take it home, and he was very excited. We hadn't even talked about it at home what that was gonna look like. So I'm approaching AI no differently from how I have really approached anything else in my nearly 20 years working in education, and that is that we should be relying on computers to do the things that are tedious for humans, and do them very quickly and very well so that we can free up our cognitive energy for the things that are creative and exciting So I can distinctly remember showing j- just an Excel spreadsheet, just I'm gonna sum these numbers quickly, and I showed, it doesn't even matter if it was a teacher or a secretary or somebody in a school, this is several years ago, and they couldn't believe how fast Excel could sum numbers. And I remember thinking like, "What do you, what do you think the computer is for?" It's, it's for doing rote calculations very, very quickly. And I wanna extrapolate that same lesson, that same moment, and extend that to what we're doing today, which is what AI should be doing is helping provide personalized feedback to students as fast and as narrowly as possible so they get information about their learning. It's the same thing that I've said for nearly two decades about getting data in the hands of people as quickly as possible so they make decisions. We took an assessment. I should know five minutes later how these kids did. What teachers were doing is getting out a highlighter, and they were, you know, building their own charts, their own histograms of their distribution of scores. Well, now that stuff is instant, right? Like, just like that I know exactly how my kids did. I should be able to turn that into action as a teacher. So now in the current era, we should be using AI to be like, "Hey, thematically, here's what I think your students didn't learn in the last lesson. Th- this, based on the pattern of scores that we have, this is what needs to be reinforced, and here's a pocket of five kids that mastered it before you taught it, and you should be extending the next thing is there," but the craft is still the teacher. The craft is still the teacher, but we're just empowering them with more information about where they should focus that creative, passionate energy. Corey: I think that there is a lot of overloading of education, uh, where you talk to some folks, they, they have a love of learning, and they, they resonate very well with formal education, and given, given their choices, they'll go back to school and get more degrees because why not? I, I, I wish I had that gift. I do not. Uh, then there are folks who view it instead as it's a credential that I need to check a box, and anything I can do to check that box faster is great. Uh, so it- it's the dichotomy between writing an essay so you can learn how to structure your thoughts and learn how to convey a point, and writing the essay so you have the essay to turn in, and you can move on to the thing you actually want to do. And AI provides an amazing shortcut for some of those things. Uh, AI detectors, honestly, they seem like varying flavors of snake oil, although we call it serpent grease now. Great. Cool. I- That's always going to be a challenge. I'm not necessarily sure how you push back on a lot of that, but I also don't necessarily think that that is the end of days. We, we saw similar freak-outs when calculators first came out. "Oh, well, now kids aren't gonna know how to do arithmetic." And yeah, there's an element of that, but also we were told, "Oh, you won't have calculators in front of you all the time." Well, joke's on you. We have them everywhere now. But also you can... Having the... Knowing how arithmetic works, so when you fumble something and get a completely ludicrous answer, you don't accept it as, "Well, it must be correct. Computer said it was." It feels like critical thinking is the skill that in many cases we are lacking as a society. But folks have been phoning it in since the dawn of time, and I don't think that that's gonna go away anytime soon. You can't shove the AI genie back in the bottle. It's here, and we have to learn to live with it on some level Adam: Yeah. I am reminded of, uh, I, I remember riding the bus when I was a kid and seeing a kid two years younger than me talking about how he got his new math books. I don't know, I was in seventh grade and he was in fifth grade or something, and I remember him saying, "Math is so easy. Did you guys realize that the answers are in the back of the book?" And he, he had just discovered, you know, he'd gotten his brand-new fifth grade textbook, whatever it was. I remember thinking like, "Okay, but y- uh, we all know that that's not the point, right? The answer to that particular problem is not the point. It's about process. If you start looking up those answers all the time, you'll never build process to be able to do the more advanced stuff." And I didn't even know what the advanced stuff was at that time, but I knew that it was coming. What you said about writing is really interesting because writing for most people is a means to an end. It's not like they're not using writing to, to... You know, they're not writers. They're not writing novels or whatever. They're writing because it articulates a point, or they're writing because they use it to help formulate their thoughts or whatever. They're, they're trying to persuade somebody. And I think that if we offload that critical thinking to a bot that then produces the outcome We are doing ourselves a gigantic disservice, and I think we will l- like, whatever hope we have right now for critical thinking developing in kids is going to be gone if we don't coach them that it's not okay to phone all of that critical thinking in, because you won't be able to write two sentences together, and you won't be able to be an informed, like, literate, productive citizen in any way. And I don't know that tools can do that. I think it has to be conversations. I think it has to be, "Listen, I know those tools are available, but, like, let's not use them at every possible step here, because you won't develop those thinking skills that are going to serve you best, um, as you become an older kid and a- as a young adult." Corey: Part of the thing I've learned as a parent is that as soon as you get used to something, like I finally got my head around this phase, there's a new one right behind it, and there always is going to be. It, it reminds me of being a technologist. This, this is not the field to be in in the event that you aren't prepared for sweeping changes. So that's a weird one. Adam: Yeah. It's ... You know, the only constant in those things is change, and so it's not about mastering something or getting past it. It's always about process. I mean, every- everything is about process in both of those fields, I think. Corey: Okay. So at this point, what do you find is the most, I guess, misunderstood aspect of tech a- as applied to the K-12 space? And you're, you're building in the space. I, I am coming at it from the other end of it, where I'm a parent, and I know how the computer stuff works, mostly when the district sends out something that clearly is this generation's version of a fax machine, which I'm pretty sure is a fax machine. This episode is sponsored by my own company, Duckbill. Having trouble with your AWS bill? Perhaps it's time to renegotiate a contract with them. Maybe you're just wondering how to predict what's going on in the wide world of AWS. Well, that's where Duckbill comes in to help. Remember, you can't duck the Duckbill bill, which I am reliably informed by my business partner is absolutely not our motto. To learn more, visit duckbillhq.com Adam: So I, I think a lot of times that I run into somebody and I say, "Oh, I work in K-12 technology," they instantly go to curriculum. They're like, "Oh, well where ... Are you using AI to replace teachers?" I'm like, I- that didn't even factor into what I was thinking about. And I think that a lot of people hear what AI and data and tech can do, and they think about replacing people and reducing costs and economies of scale, all those kinds of things. Those don't even factor into what I'm doing. I, I literally never think about replacing people. It's about improving efficiency so the people that still have brains are actually more able to do the work they want to do effectively, the part that as of right now AI can't do, or if it does it, it does it poorly. I think that the misconception is that everyone who works in tech in the K-12 space has that bent, that everyone wants to do that, and that's, that's not the work that I am doing. I am largely doing integrative work. I'm trying to combine data from multiple sources into one place so that we're not spending a lot of time jumping between reports. I, I think that's the, the biggest misconception that people probably have about what we're doing in this area. Corey: Where, where do you find the good vendors in this space? Because I've, I've worked with some in the past that I thought were decent, and then invariably I start looking and, I'm an active parent, I start seeing what is being pushed to kids and the rest. Like it's, it's weird now, where I log into the SFUSD district, uh, website for, they have a whole portal and they use an app. Uh, I think it's ParentSquare combined with ParentView. Cool. And they apparently have Google Workspaces for my kids. An email address has been set up for them. Okay, cool. I don't recall ever encountering a data privacy addendum for this, uh, the standard terms that Google Workspace has. I have some challenges as applied to kids. Um, it, it feels like I, I ... On the one hand, I feel like I'm a relic and that I'm the only one that seems to care about this stuff anymore, but also these are kids. We all did stupid stuff when we're kids. One of the shining graces of being of our generation is that we didn't have a permanent digital record being tied to us every time we screwed something up. They used to threaten us with, "Oh, this will go on your permanent record." By the time you're an adult, you realize that doesn't exist. Uh, it turns out that my, my, uh, when I apply for a job, the new employer can't pull up how many times I was given detention in middle school. Adam: Thankfully. Corey: But now, who knows? Maybe they can. Adam: It's somewhere. It's probably leaked already. That's the thing. It's not even on purpose released. It's just leaked out to the internet through some, you know, one of those, uh, you know, one of those big hacks. Yeah. So a lot of the vendors-- Your original question is, like, what, which, how do you find the good vendors? So ones that take data privacy seriously is pretty important, so they will right away when you start talking with them, like we have, you know, we've taken this pledge, um, to make sure that data is protected. Um, here's our procedures that we have in place. I mean, ultimately, almost everybody at some point gets hacked. I just want to know that they're not really loosey-goosey with the data and they're, they're taking that part seriously. They're not serving any ads to kids. There, there's no in, you know, in-app purchases. If, if it's free to the schools, that means they're monetizing in some other way. That always sets off an alarm bell. So when some, some vendor is like, "Oh, hey, um, you know, it's, it's free to try out, and, and this is how this works," that always, that always worries me. Like, I-- You're-- Some-somewhere you're taking that data and you're selling it to someone. That's the only way you can be profitable. But then I think it really comes down to what do the teachers and the kids like about it when they see it? So if we pilot something with teachers and they get excited about how creative they can be in it, I'll come back to that creativity, just that idea of I know that I can build an experience for the kids here that is going to be meaningful and actually teach them a skill or help them understand a concept. If they see that and get stoked about it, then things like adoption are gonna be really high because they can actually see themselves using it. When, when that stuff happens, then I, I know they're excited. And then on the back end, if I know they're gonna help us integrate, like I'm gonna be able to get roster data and build the classes for the teachers and actually capture some of that information about their learning, that's incredibly important for me as the integrationist. Corey: I wanna get back to that creativity piece you just mentioned. Part of the challenge too with this is that I, I am a relatively sophisticated consumer because I have an entire decade of being professionally cynical as companies try to pull fast ones. Uh, I don't know necessarily that-- I, I don't even wanna single out teachers on this, like let alone the kids. I don't think the general population is well equipped for this. Most folks do not have the wherewithal to s-read, actually read the terms of service that are voluminous, uh, and find the hidden weasel phrase in there. They don't know to inspect, uh, what trackers are on this site. They don't have the bevy of extensions they do for browsing the web, uh, and not winding up infected constantly that most technologists have, have sort of learned over the years. I, I, I find that that is-- it is a skill that increasingly feels rarefied. Uh, also not for nothing, so does being able to intelligently tease an answer out of Google when punching things into it. You wouldn't think that would've been a durable skill for 20 years, and yet here we are. Adam: I don't know that I am as diligent in reading those things as I should be as a per- as a personal consumer. When it involves the students, I, yes, 'cause it's somebody else's, mm, you know, wellbeing that I'm worried about. When it's mine, probably not so much. I will say that one of the, the very first uses of AI that I thought of when I learned about what, like, ChatGPT was able to do, I said, "I mean, we should feed every terms of service into this and have it start identifying the things that we should be worried about." Or other ... Like, I sign contracts every now and again with school districts or with vendors. I'm like, I should be able to run this through here and say, like, "Hey, these are the things I normally care about. Do you, AI agent, do you find anything in here that I find objectionable?" Um, I never, like, built anything, but I No, I suggested it to a few friends. Like, if you're looking for something to build, maybe it should be this if you have expertise in this area. I- Like I say, I don't know that I am as diligent as I should be about my own footprint because I don't know that I'm ever really creating a lot of data that I'm very worried about other people getting their hands on. So I guess I don't, I probably don't worry about it like I should. Corey: What do you think right now people are sleeping on with respect to, again, the moment that we're in, ideally through the, through the lens of K through 12, but I am curious about the larger picture here. Adam: So I don't know how much you read about this or how much you hear, um, in the space that people are building tons of processes that rely on AI, and the, it's funded on speculation, and that at some point the costs are going to go up because the, the companies will demand to be profitable. Corey: I, I, I might have heard some of this, yes. Adam: Okay. Right? Like, like, think about the, the rug pull that that will be in K12. Because you mentioned it earlier, uh, you know, I was really excited about Google Jamboard, and suddenly that's not a thing that I can use anymore. And imagine that it's not just your Jamboard, it's everything that you have. Imagine that suddenly everything now is the costs are astronomical to the point that you can't afford doing any of it, and we've built a house of cards of, of stacks of educational tools that now suddenly either they go away entirely because the company's not profitable, or it becomes priced out of what, what a school can pay for, and it just, we... You know, maybe that's five years away or something. We'll have forgotten how to do the other work w- without it, and that's a scary thing. Corey: I have been as- astonished that no one else seems to have successfully followed the Apple play back in the '80s and '90s, where every computer lab that I ever went through, K through 12, had Apple everywhere because they gave huge discounts into the education market. And wouldn't you know, there's a reason now that if you go into most developer environments, uh, in mo- at least in the startup world, Macs are everywhere. It's because it's what people used in school, they used growing up. They used Windows, realized it was terrible, and, "Okay, I'm gonna use the thing I already know how to use." I, I was-- I, I shout the thought it was blue in the face that AWS should have gotten into every school, every boot camp they could find with mountains of free credits, and they didn't, and now here we are. Adam: I think that that, the, the chickens coming home to roost moment, I don't know how far away that is, but I think it's, you know, it's built on a gigantic pile of cash that someday will run out. And instead of ensuring that there's the long... I, you know, I don't know if this was the master plan for Apple, get it into the hands of kids so that when they become adults, they'll use the tools. May- maybe. But, like, is anybody doing that play today? Probably not. And they should be because otherwise what, what's going to happen when there comes a time when someone has to make decisions about what tools they're using? That decision will become way harder than it could have been if it was subsidized as part of the way you're describing. Corey: I will say that apparently we already covered once in this episode already that Google Workspace is basically there playing that game. I'm sure the pricing is attractive. And okay, I, I will at least concede that it is unlikely that they're gonna kill Google Workspace by the time my kids graduate high school. Not impossible, they're Google, but unlikely But yeah, you don't see a lot of the rest of it. And especially with this, you're right, with the AI stuff being thrown in, when, when this switches to a per token cost and every usage of AI becomes an investment decision, are we really gonna want it all the places it's currently in? Adam: Probably not. I will say the one thing about Google that they probably did right is that when, when COVID hit, they were positioned so well with Chromebooks. Because I remember 10 years ago when we were like, "How are we gonna afford devices for kids?" 'Cause a laptop was $1,000 a student. There's just no way we're gonna be able to have devices for s- for kids. And then when Chromebooks came out, and they're like, "Well, you can get one for 200, 250, $300 with a warranty," you know, a little bit more than that, and then it's bumper-to-bumper coverage. When COVID hit and everyone's like, "Oh man, we need to get devices for every single kid," I- everybody bought Chromebooks. They were back ordered for, you know, months and months and months. And I am now seeing it in real life. I have a nine-year-old who doesn't really know how to use Windows. He is comfortable on a Chromebook, that's what he knows, and when I show him the Start menu on a Surface laptop, he does- he's all thumbs. He has no idea what to do 'cause he's never used a real actual computer. He knows Chrome OS. I don't know businesses that are using Chrome OS. I don't know that that will ever really happen, that everyone's adopted that. But it certainly, they got them into the, they got the devices into the hands of the kids, and now they're in the Google ecosystem for that, for that exact reason. Corey: Uh, I do have one further question for you. Uh, you've been using AWS for a while in interesting and strange ways. Do you still find that it's, or I guess, do you at all find that it is the thing that you reach for when you're building something new? Like, how, how are you engaging with it? How are you using it? How have you seen it change? Adam: It is the thing I'm using, and I don't reach for much other tools. Um, I recognize that the landscape has changed, and people have created very niche products out there that do a very specific thing better and cheaper and faster and easier than AWS does, but I'm in that system now. And so how I got into it- Like eight-ish years ago was I had a, a task that needed to run for a bunch of schools, and I very quickly outstripped the capacity that like a single server could have, like one, one server. And I had heard about cloud computing, and I kinda knew that was probably my answer, and I did a little bit of reading, and I, oh, if I, if I containerize this work, I can stand up a cluster of computers in seconds and have it do all this work. I very, very quickly, over the span of a couple of weeks and months, learned how to stand those things up, and now I just shop for problems. 'Cause now that I can build a stack in minutes and do a lot of work really quickly, that's just the place that I build in. I will use AWS until it doesn't meet my needs, but I read all the time about, oh, this is faster for deploying a, you know, a little, a little app, and this is faster for doing this, and this might be faster for developing a, a, a CDN that serves content. Ultimately, I don't really care as long as what I'm building isn't f- you know, 100 times as expensive as it would be on some other thing. Uh, yeah, I, I'm not excited every time I have to spin up a new NAT gateway because- Corey: Imagine that ... ' Adam: cause we're just lighting money on fire to move data with, you know, within my own network. That's just life. Um, but you know what? Fun, Corey: isn't it? Adam: It, it sure is. Okay, every one of those now is, every project is its own $35 a month spend. Okay. But it works. So like because I have a stack that works, and every new problem that comes to me, I'm like, well, I already have a thing that I can just spin up and then route different work into that one, I'm gonna keep doing it. That's, I basically am just shopping for the next ETL task that some K-12 place needs, and I'm gonna zip data over to, to that school or zip data out of that school and do something with it. Um, that's the problems I'm searching for and continually implementing solutions for. Corey: I really wanna thank you for taking the time to speak with me. If people wanna learn more, where's the best place for them to find you? Adam: I am auroraedtech.com for the company, and then in every usual place you find people, I am aplarsen, A-P-L-A-R-S-E-N. Corey: We've, of course, put links to that in the show notes. Adam: Excellent. Thank you so much, Corey. Corey: Adam, thank you so much for taking the time to speak with me. I appreciate it. Adam Larsen, principal developer at Aurora Ed Tech. I'm Corey Quinn. This is Screaming in the Cloud. Stick around. View Full Transcript Hide Full Transcript ## You might also like [More Podcast Episodes](https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/) ### [When AI Starts Writing the Pull Requests with Madelyn Olson](https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/when-ai-starts-writing-the-pull-requests-with-madelyn-olson/) Screaming in the Cloud 06.25.2026 30 Minutes [Play Episode](https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/when-ai-starts-writing-the-pull-requests-with-madelyn-olson/) ### [The Appalachian Cloud Trail: Hiking, Cloud Economics, and Finding Perspective](https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/the-appalachian-cloud-trail-hiking-cloud-economics-and-finding-perspective/) Screaming in the Cloud 06.11.2026 33 Minutes [Play Episode](https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/the-appalachian-cloud-trail-hiking-cloud-economics-and-finding-perspective/) ### [Coding Agents, Chaos, and the Future of Dev Work with Dexter Horthy](https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/coding-agents-chaos-and-the-future-of-dev-work-with-dexter-horthy/) Screaming in the Cloud 05.28.2026 33 Minutes [Play Episode](https://www.lastweekinaws.com/podcast/screaming-in-the-cloud/coding-agents-chaos-and-the-future-of-dev-work-with-dexter-horthy/)