44: AI, Burnout, and the Myth of the 10x Developer: Addressing Burnout in Software Engineering

Erika (00:01) Welcome to the Overcommitted Podcast, your weekly dose of real engineering conversations. I’m your host this week, Erika, and I’m joined by…

Bethany (00:10) Hey, I’m Bethany.

Brittany Ellich (00:12) and I’m Brittany Ellich.

Erika (00:13) The three of us met working on a team at GitHub and quickly realized we are all obsessed with getting better at what we do.

So we decided to start this podcast to share what we’ve learned. We’ll be talking about everything from leveling up your technical skills to navigating your professional development, all with the goal of creating a community where engineers can learn and connect. Today on Overcommitted, we are opening up about what may be contributing to a sense of burnout and how we can balance our ambitions with the pressures of 10x development and an always on working

According to the numbers, our industry is facing a paradox. The opportunity to work from home and use AI assistance promise increased productivity and work-life balance, but data reported by Lead Dove reveals that 66 % of tech workers report experiencing symptoms of burnout, with the staggering 22 % facing critical levels, a figure that now exceeds

even the height of the COVID-19 pandemic. For our discussion, we’ll define burnout as a state of exhaustion, cynicism, and reduced efficiency caused by prolonged stressors, including work pressures and intrinsic factors such as mental stress or job insecurity.

So let’s start with, do you feel like you’re facing some level of burnout? And if so, what are some of the major contributing factors?

Bethany (01:48) For me personally, I don’t think I’m facing burnout. think, like…

On my team, I’ve been able to strike a good balance of having like times where I’m pushing and then times where I’m like more reclaiming that energy. And I feel like I’ve been able to communicate that a lot better. But I have definitely encountered like burnout conditions previously in my career. And I think it’s taken a lot of confidence and having to get over a lot of that imposter syndrome to

have that ability to say no, to push back and say, okay, I’ve done my pushing. I’m like, now I’m going to take it easy for a bit and reclaim energy because I mean, that’s such, it’s like how you get a good ergonomic keyboard for your wrists and to protect your wrists. Like you also have to protect your brain too and protect your sanity.

Erika (02:41) So it sounds like you’ve experienced burnout in the past, but at this current moment, you’re not feeling a burnout.

Brittany Ellich (02:41) I love that so much.

Bethany (02:49) Yeah, definitely, definitely.

Erika (02:51) Well, that’s good. can help inform some of those things that you do that you’ve learned that can help prevent burnout for others. What about you, Brittany?

Brittany Ellich (03:04) Yeah, I guess it’s kind of hard to say exactly. I think that I might be approaching and flirting with burnout occasionally and trying to really recognize, all right, when does my mental health need a break and stuff and trying to take those breaks. And I think that most of the contributing factors there are factors that I put upon myself more so than ones that are coming from work.

because I have a lot of things that I want to do and I get really excited about doing those things and then I have too many things happening at once. I think at work, there’s definitely some stressors. mean, there’s like a constant fear of, you know, layoffs or, you know, things changing massively at work that are part of the reason why I do a lot of different things. So I think that like the…

don’t if it’s necessarily peer pressure at work, like the, you know, the fear of that, I’m like, oh, okay, I should be doing more stuff so that, you know, if layoffs do come around, hopefully I’m not considered because I can add all this extra value to the company. So like that’s a contributing factor that’s at work, but not necessarily like directly related to my exact job. Um, my current manager, uh, one of the reasons why I will stay with the team that I’m on for as long as I possibly can is like, she’s just incredible and really good about.

I’m like, okay, but how are you doing? Are you okay? And I really, really appreciate that. And yeah, she’s one of the best managers I’ve had. think that that helps a ton. But also like just like the world right now is crazy. And so like, I feel like I’m experiencing burnout from just like, my God, the news, the news that is happening is just so much. ⁓

Bethany (04:46) such

a valid point, yeah.

Erika (04:49) Yeah, yeah, I mean, it’s definitely true that there’s like things that we can do and like control and help, but there’s also things that are outside of our control that are these contributing factors and can kind of take you over the edge. If you’re at that kind of like level of stress that feels like somewhat manageable and then something happens that like pushes you over the edge.

I definitely, I’m with you. I feel like I kind of go in and out of that state of feeling close to burnout. Like the big indicator for me is the reduced efficiency.

I’ve definitely had those days where I like sit down and I feel like I go through the whole day and I get nothing done because I can’t find the motivation. And like that’s usually a sign to me that something’s not quite right because usually like in my current moment I feel motivated. I like feel like I can…

I can approach the start of the day and like, you know, take on the work and like actually get it done. But yeah, I think you’re talking about kind of the human factor and like having people check in with you and like, yeah, I do definitely feel like that’s a big factor for me of like feeling safe and feeling like I have

people who care about me. Yeah, like it’s it’s almost never really about the work for me. Like I, I don’t, I don’t care as much like what specific tasks I’m doing. Like certain things definitely do energize me more than others. Like a lot of the like managerial stuff does not excite me as much, but

Yeah, really I feel like it’s the connection and feeling like I’m valued, whether that’s me telling myself, like, this is valuable. Like the work that you’re doing is, you know, having a good impact or somebody else telling me that, yeah, I’m valued and I’m doing good work.

Yeah, well, according again to the data and some industry trends, and none of us mentioned this in what we have cited as maybe indicators or contributors to our burnout, but maybe it’s because it wasn’t top of mind and this is kind of new that evaluating the impact of AI on the

effects of burnout and the experience of burnout, there have been causal links found. And, you know, last year, a lot of these studies happened of like, is AI making us more effective? And some of this pressure for that, we’ve heard like the 10x developer come from this idea that

hey, now that we have AI assistance, we can be 10 times more productive. And a lot of those studies have not really proved that out. So there’s a group called Model Evaluation and Threat Research, which is a nonprofit dedicated to evaluating AI models and helping us evaluate their

capabilities and risks. they found that in rigorous trials, a lot of these with experienced expert open source developers using AI for complex tasks took actually 19 % longer than those who worked without them. And other studies have found from various security groups that

Since AI models are trained on historical data, they often reproduce outdated patterns or vulnerabilities. studies have found that up to 30 % of AI-generated snippets contain security issues like SQL injection or cross-site scripting, which were previously considered solve problems in most modern frameworks. So as we know, AI is not magic.

and requires thoughtful context management and code review to use well in software projects. So I guess, first of all, does any of this kind of ring true to some of your own experience? Has AI kind of shifted the way that you approach your work as far as creating code versus mastering context?

Brittany Ellich (09:24) Yeah, I think that this does make a lot of sense. I’ve noticed recently that there’s sort of a divide in AI tools that are out there where there’s like the synchronous tools like code completions and chat, I guess is pretty synchronous because you’re asking questions and getting stuff back. And then there’s the asynchronous tools that are like the agents where you’re delegating tasks to them. And I feel like the synchronous tools were really easy to adopt and didn’t have a huge learning curve because there were already a ton of things like it was repeating.

existing patterns, know, like resharper existed for a long time, which also had that like tab thing or whatever. So there wasn’t, there wasn’t like a huge learning curve to take them on. But now as I’m trying to find, you know, pursue the 10x developer myth, which I think is definitely a myth. Cause that even the idea of being 10 times more productive gives me anxiety. But,

As I’m doing that, I’m realizing like there’s actually a huge learning curve to learning how to delegate those things out. And so it makes sense that it takes longer to do a lot of those things, because I think it’s really easy to like use AI tools, but to use them well is really, it takes a lot of time investment to get to the point where like, you you’ve asked yourself, could this be done with AI or not? Enough times that you, you you know what the limitations are and what the…

you what is faster if you do it yourself versus what’s faster if you give it to AI and yeah, all the context switching that goes along with it, which is also, you know, problematic and hard to get through. So yeah, I could say that that…

It brings true for sure. I’m sure at some point we’ll get to the point where things it does actually make people faster. And I think in certain contexts, things like code completions almost certainly make people faster because it’s not there’s not a learning curve to it. feel like as there is compared to like other tools, but

Yeah, as many things, it depends. My favorite answer.

Bethany (11:09) Yeah,

it really does though. I agree completely. I think it’s a shift in how we’ve traditionally worked. I think with things taking longer, it certainly does while you’re learning how to use things. But I think another thing I’m encountering is just like…

having a disconnect from what code is actually being produced because I’m not necessarily the one typing it or coming up with the idea or like going forward from there. And so I think that’s really the toughest part for me. And I’m very interested with like, I think they’re Anthropic recently said that cloud code is 100 % just they use a agentic like cloud.

quad code to build quad code. And that fascinates me so much because it’s just tough to feel engaged when you’re saying, okay, generate this, generate this, generate this. not like you’re reviewing the code, but it’s, there’s still like that, that disconnect. And I think that’s the piece that I’m really trying to like rediscover as like part of learning this tool set, because it’s so hard to be like, I mean, we’re

We’re engineers, we signed up to solve problems and it’s hard when you’re not necessarily the one solving the problem anymore. You’re just delegating it to something else to solve that problem. And I’m sure that’s like been a historical issue, like managers probably encounter the same thing, but they’re solving problems in different respects. it’s definitely an interesting space right now.

Brittany Ellich (12:42) It definitely does feel like almost like a managerial role that you step into because you’re delegating all of these tasks out. I think it’s also interesting because it’s sort of changed the way that code review happens because like when I review a code from my colleague, I’m like, I know that you’ve like been testing this along the way. You know how this works. Like I’m going to read it, but I’m not going to like go in and like be upset about different stylistic choices that you’ve made. Like if things work and there’s nothing like egregious about it, I’ll probably approve it. But like I have to have a much.

deeper level of review when it’s completely generated by AI, because it’s still attached to my name. It’s still the code that I wrote, quote unquote, and I’m responsible for. But did it test it as I went along like I would if I wrote it myself. And so there’s definitely more cognitive load associated with the review, I think, than there would be otherwise.

Erika (13:33) Yeah, yeah, I think we haven’t had this experience, but I’ve been reading about people who have gotten code reviews or like PR review requests and the person who wrote it takes very little responsibility for the quality of the code. And, you know, I mean, that’s kind of a human problem, not really like an AI problem. You’re like, well, would that have changed if they had written the code themselves?

or not but yeah it’s definitely something to watch out for and even in your own like you can’t control what other other people do but you know in your own PRs like taking that same approach of like looking at it almost as an outsider because I guess you kind of are right like you are reviewing the code that got written and

Yeah, like going through it with a microscope and recognizing that still does take.

like you said, a good amount of cognitive load to do that. Like peer reviews are not free. They’re not easy necessarily. You’re using all the same muscles that you would if you were writing the code. And it can actually take longer, which might be where this, you know, 19 % decrease in productivity comes. Like it is that ramp up factor of

learning how to use the tools effectively, but then also there’s the idea that it can take longer to read code and understand it than to write it yourself. So yeah, I’m with you. The 10x developer myth, mean…

Yeah, probably like amount of code that’s created can be 10x, but the amount of that that actually gets merged and doesn’t have to get reworks or churns, I have never come anywhere near that. At least for production code, know, vibe coding or like Greenfield hackathon projects is one thing, but for production code, definitely not.

And yeah, this idea of context mastery too, know, context mastery and then like the sort of context management. It’s, yeah, it’s an interesting, again, mindset shift and something that like I feel like would be interesting to kind of like measure for myself. Like…

how long it feels like it takes me to master some set of context. And then also like share that with some kind of generating model. And yeah, I guess I haven’t really found that…

I guess that point where I feel like the context that I’m giving to the model really meaningfully changed the output that I get. But I know, Brittany, you’ve been doing some testing with benchmarking your code. And Bethany, I’m guessing you’ve done some of this with Copilot too, of benchmarking.

results compared with different inputs. So you might you both probably have more experience with this than I do.

Bethany (17:00) Before going on, I’m curious how you define context mastery, ⁓ especially in today’s world or whatnot. Is it just knowing the context or is it truly having a large handle on it?

Erika (17:05) Yeah.

Yeah, I mean, it’s a good question. And it also has two meanings where it’s like my personal context or like my personal understanding and then the like context that I give to the AI. But like the probably the meeting, the meeting point is like, I feel like I have mastery of context when I can explain it to somebody.

And that’s probably also the point when I can tell an AI what to do.

Brittany Ellich (17:48) Gotcha. That makes sense. think that there is, there was a article that I read the other day that talked about like how prompt engineering is no longer a thing and now it’s all about context. Like if you give it enough context, then it’s going to be good enough to get the thing done. I can’t say that I’ve spent a ton of time like really trying to optimize for prompt engineering or for context. I always go by the

the vibe basically, this is vibe coding, but not actually like the term vibe coding. I’m like the vibe of like, if I gave this to somebody who was brand new to the code base, is this enough information for them to do the thing that I want them to do? And that’s always been enough for me to like get pretty decent results. Sometimes I realize like the things that typically goes wrong is, you know, I skipped something that was important or I gave it too much at once. But.

Typically like thinking like, all right, this person is brand new to the code base. If I were writing this for like an intern or somebody brand new on our team or something like how, what would I need to provide? And that almost ends up being enough. I can’t say that I’ve done a ton of like testing or, you know, tweaking of that over time, other than like just doing it and figuring out what’s too big and what’s too little.

Bethany (19:02) Yeah, I think for me, it’s really been about having a plan step in there. it’s really about being explicit with what you want and like Brittany and Erica, you’re saying providing the correct context and patterns that you want it to follow. And so I find it valuable to always take that extra time to have it plan what it’s doing because I…

I have not enough patience to prompt engineer and describe it in the way that it needs. And models change so frequently that I think it’s impossible to keep up with that standard. And so I think largely having it do its own planning and then being the reviewer for that has made it a little easier to kind of guide it to what I want. But I still…

feel like there is always some level of reworking you have to do at the end. Unless it’s like super tiny slice of work, it’s like rare that it gets it right the first try.

Erika (20:01) That’s good. Thanks for sharing some good tips. Well, great. Let’s shift the conversation to some ideas of boundaries. Bethany, you of mentioned this at the beginning of setting boundaries as something that’s helped you. And with this idea of the pressures of 10x development,

I thankfully also have not felt the pressure to deliver faster necessarily in the age of AI, but kind of found some scripts that I wanted to share if anyone is kind of facing these, I think hearing somebody else say it or having something prepared is really helpful. So if anyone is kind of facing these,

heard some no scripts, a way to approach that conversation healthfully is instead of saying, I can’t do this, maybe you’re getting, yeah, somebody’s asking you to ship something that’s not ready, ship something that hasn’t been fully reviewed, you can say,

something along the lines of to deliver with architectural integrity, I need to de-prioritize. If we add this, we risk the stability of development and the deployment pipeline. What trade-off do you prefer? So that’s kind of that managing up piece. Or saying something like, I am slowing down this sprint to ensure we don’t spend the next three sprints fixing AI-generated bugs. So.

Those are things you could try if you’re under that pressure.

Brittany Ellich (21:37) one thing to add there actually. I feel like this is something that I’ve been asked to do probably over the last year or so as I’m working with some teams that do work very, very quickly and want quick turnarounds. And I think that one thing that I found really useful is to like try to put things in the terms of, you know, whoever it is.

that is making the ask. Like a lot of times it’s a product manager saying, hey, we need this. And then saying like, all right, which is the priority, this thing or this thing? Because we can only get one of those things done. And I feel like that has been really useful to level set and be like, all right, we can only do so many things. You pick which one you want me to work on. And I’m happy to do that. But we can’t do both. And I think being really

honest and upfront about that is just better for everybody, just to level set expectations and also to, you know, actually deliver a working thing at the end. There’s still a lot of things that end up getting deprioritized, you know, after we had already started working on it. But I think that’s just the nature of work right now. Like everything’s moving so fast that sometimes we realize, we don’t need that. Let’s move on to something else. And it’s harder because I take, you know,

Like a lot of engineers, I feel like I take a lot of pride in the stuff that I work on and I like want to finish that thing if I’m working on it and want to get it all the way to completion. But sometimes I think we also have to accept that, you know, that’s just not what the product needs right now. And we need to move on and do something else. And it’s a skill to build for sure.

Bethany (23:07) Yeah, I definitely agree that.

communicating trade-offs and being able to make it so that it feels like the person who’s asking has some contribution to what gets done and what happens. But I think it’s also important to note that there’s some level of psychological safety that’s needed to push back and to say no. And I know we’ve had an episode on psychological safety and the importance of that, but it really is

like that I think first like foundation level to be able to push back to communicate and for everyone to have the trust in each other that we’re making the best decisions with what we have and trying to do the best we can. And so I would say if like, if you’re listening to this and you’re like, wow, I don’t feel like I could do this. Like I think that’s probably a problem you should solve before like figuring out how to push back is like.

Is this a culture issue? Is this a imposter syndrome issue? Is this like, why do feel like you can’t push back? And then fixing that first before going on with like, okay, now I can push back and like have a real conversation about expectations here.

Erika (24:15) Yeah, something I have trouble saying no to is this idea of glue work. And I’m hoping that maybe you two can help me with some things that I can use to say no. And there’s now some conversation of like, hey, maybe in this age of AI development, maybe glue work isn’t so bad.

you know, we’re all leaning a little bit more into, if anyone’s not familiar with the term, it’s sort of this less technical side of the job, like helping unblock team members, reviewing design documents, coordinating tasks. And in the past, for individual contributors and engineers, there’s been this idea that this doesn’t actually contribute to your career progression.

So, but I feel like to your point, like I get asked to do this stuff a lot because I don’t know, I’m fine at it, maybe because I’ve said yes a lot in the past. And I think even when it might be more valuable now.

there’s this idea of documenting all the glue work that you do and making sure that it’s valuable and actually delivering some kind of measurable positive outcome, but also saying no to the stuff that is not technically aligned. And I’m curious if either of you have experienced this. I’m guessing yes, because it’s…

It feels like a very common thing, especially for women to get asked to do this kind of work. And yeah, how do you prioritize and what do you say when something’s not in that area of something that’s gonna help you move forward in your career?

Bethany (26:15) Yeah, I think I definitely have been been like part of that. And I’ve also leaned into that glue work because it feels like something that I can meaningfully do well when I’m afraid of doing maybe more technical work and being seen as like a fraud or or whatnot. So I think like, definitely some early I this isn’t to say there is not an issue in the industry of like stuff.

especially women and younger engineers getting assigned glue work and it not contributing to career progression just saying that, but I’ve also leaned into that. So I think it’s like worthwhile to have people who are willing to push back on that. I would say that I just started communicating with my manager the things that I needed and just being upfront saying, hey, I’m

my goal is to get promoted and I really need to show X, Y, Z to get promoted. Therefore, I need to get these opportunities to do that work. And just being clear with whoever’s assigning your work or assigning initiatives, like what you need, which is unfortunate. Like they should also be able to know that, but it is good to be extra clear on that. And then if you’re asked things, like on my team, we’re asked for a lot of like,

quick things or ad hoc things that may or may not be important. And I think it’s what I tend to do is say, hey, thanks for the suggestion. Let me create an issue and I’ll check on the priority here. usually delegate that to people who are actually like a siding priority and actually like in charge of that, especially if it’s something where I’m like, okay, this isn’t going to meaningfully help me. And I don’t know that it’s really important for the product. So.

Let’s actually check with people who have more say there rather than just going on with the ask.

Brittany Ellich (27:57) Yeah, I think that sort of what Bethany was alluding to, like the prioritization of the work, like if it is actually a priority and will help the team, I think that it’s probably actually very beneficial to your career.

And like you said, Erica, I think it’s probably like becoming more important over time. And I think that people are starting to recognize that like, yeah, this work is important. Like it’s better to make your entire team go faster than for you to go faster as an individual. If you can level up your entire team, that’s much more impactful than, you know, just getting really good at doing stuff on your own. And I feel like that’s sort of the role of like, you know, more higher and higher individual contributors now involves a lot more glue work.

I think, yeah, just recognizing when something’s actually gonna be useful to the team and to the product and then when it’s not and pushing back on that is important. And like Bethany said, I love the idea of telling your manager, like, this is my goal, this is what I want. And then hopefully they’re giving you the opportunities that will actually contribute to that and not necessarily just throwing things at you.

And just don’t offer to take notes, unless you know, unless you really want to take notes so I feel like that’s the one can you take notes. hey that’s true that’s a good point maybe some of the glue work yeah maybe some of the work is actually like really good Ai things ⁓ yeah.

Bethany (29:05) There’s AI for that now. Anyways, use AI.

Yeah,

definitely. Yeah, just be selfish. Like, ruthlessly do things that benefit you.

Brittany Ellich (29:25) Yes, that’s hard. It’s hard to do.

Bethany (29:26) it.

Yeah.

Erika (29:29) Yeah, yeah. I, yeah, so the idea of like creating alignment with your manager and having the same priorities of what is and is not important for you personally and then also for the team. Yeah, that.

that creates a team within your team to look out for your career. And ideally, your manager respects that and understands that and believes in you enough to push that forward. yeah, and also reducing the urgency of those asks by

making it asynchronous, but making it public and visible, making it so that somebody else can pick it up. Yeah, those are all really great suggestions. Yeah, and thank you. Thank you for your input and helping. Yeah.

good to look out for ourselves and ultimately like you said it benefits everybody when the most important things are being worked on and also when we’re empowered to do our best work. Yeah, because if it’s not the most important thing, if it’s something that can wait or something that can be done by somebody else maybe more

quickly or something like that or by AI. Great. Like, yeah, it moves everyone forward. So awesome. Well, thank you both for engaging in this conversation. Any final thoughts on this topic before we wrap up?

Bethany (31:09) I don’t think so. think it’s good to always pause and like talk about burnout and normalize it because I think it’s something that people almost treat as a bad word. Like, because once it’s almost like once you name it, it’s real. But I think it’s important to check in about that when you’re not feeling it even to and just like regularly check in with your mental health, check in with yourself and be real with yourself and your team and your manager and stuff.

what’s happening. Of course if there’s that psychological safety, I think most of those conversations rely on having that with your team, but if you do, definitely talk about it more. Normalize mental health conversations.

Brittany Ellich (31:50) Yeah. Plus one to that. I am a huge fan of the fact that I feel like mental health conversations are becoming less taboo to have. and you know, trying to find what it is that works for you is great. I love that. I’m on Prozac. It’s amazing. It’s my favorite. Like, I love it. It’s great. So find the thing that works for you. If that’s medication or if it’s therapy or if it’s, know, whatever it is that you need to do to survive in this crazy world, then yeah.

Do it if you’re not feeling good.

Bethany (32:20) I think that’s such a good point. Like, like, medicine’s great, therapy’s great. Also, it’s okay to not be okay with how wild things are right now. I think it is. If things are weighing you down in the world, like, that’s normal. That’s just a sign that you are a normal individual who can feel empathy for these things and like, it’s okay.

Brittany Ellich (32:41) Yeah, yeah, if what you can do is the bare minimum to get through, like, I think that’s fine right now.

Erika (32:48) Yeah.

Yeah, and I guess also we were talking a bit about the beginning that it can come in waves and it can come in and out. You can feel more or less burned out at different points in time. yeah, even hopefully having supportive colleagues, people you can talk with if that’s helpful for you or yeah, any other any other tools in the toolbox to help prevent burnout ideally. But then if it does happen,

to recover from it. So, well great, let’s wrap up with our fun segment. Today we are making wild predictions. They need to be wild. We were talking before that they can be mediocre and I’m saying no. No mediocre predictions here.

So wild, wild predictions, the wilder the better for 2026. So I guess I can start. My wild prediction for 2026 is that by the end of this year,

overcommitted will have read five books. That’s not wild. These are so hard to make wild.

Bethany (34:07) That is such a wholesome

prediction. I love that. ⁓

Brittany Ellich (34:11) together.

Oh, yeah.

Bethany (34:13) that to

happen, yeah. I stand by that.

Erika (34:14) Yeah, I’m a man.

myself will have written 10 blog posts there that’s wild considering I have written one in my entire lifetime okay there we go crazy

Brittany Ellich (34:24) There we go. Yes. Nice.

Yeah. For those who don’t know, we do have our Writing for Developers blog or book club going on right now. And we’ve had a lot of wonderful contributions in our Discord of people sharing blog posts. So.

Erika (34:39) Yes.

been so fun.

Brittany Ellich (34:42) It’s been great. Wow, I don’t know how to follow that up. That’s a… I think I’ve been thinking about this a little bit. think that my wild prediction is that there’s going to be something that exists to help manage the context switching for AI, like asynchronous AI development. I don’t feel like that tool exists yet.

Erika (34:49) Bye.

Brittany Ellich (35:10) and like, feel like my brain breaks far sooner than, than the, then the agents break, but that my wild prediction is that we’re going to have figured that out by the end of 2026. hope, I’ll share one article, in the show notes about that, that I think was kind of interesting from, Gastown where they were talking about like, you know, agents kicking off other agents and

handling other agents, feel like there’s going to have to be something similar to that. But as we’ve already said in this conversation, that might not work in a production environment where you actually care about the code afterwards. hopefully there will be something that exists that handles that.

Bethany (35:50) Gosh, that’s so real. There was a thread from the person who made Cloud Code on how they use Cloud Code, and they said something about kicking off five instances of at a time, and I’m just like, how do you prevent it from conflicting with itself? How? So hopefully, hopefully that’s something that people are willing to share or help out with.

Okay, my aim is I think that there will be an uptick in like bad web design to almost like prove that something wasn’t done by AI, like to almost like revolt against that. So I think like, like the GeoCities like things like that, that’s going to come back and and like be present to almost prove that something was not like

like done by AI is my prediction.

Brittany Ellich (36:48) I love that so much. feel like there’s been several things recently where I like actually wrote it out and I almost wanted to say, by the way, this wasn’t written by AI just to like give it a little bit more credibility. And like I’m somebody who’s very pro like using AI for everything. So I can see that I’ve noticed like I read a blog post yesterday and I always noticed typos and now like AI really tools don’t really produce typos. I was like, sometimes I’ll go and tell the author like, by the way, you spelled this wrong. Now I’m like, nope, leave it.

evidence that this was a human that wrote it and that’s not a bad thing. grammar.

Bethany (37:20) Yeah, 100%. Every time it adds an dash, I’m like, nope, we’re stripping that out. No dashes in my written work.

Erika (37:28) I felt called out because I was talking to somebody the other week and she, I guess she’s a teacher and she said that like anything she encounters in her work like that has emojis in it was written by AI. I was like, ooh, I use emojis all over the place. This is like one of those like, ⁓ am I AI? Uh-oh.

Brittany Ellich (37:49) No!

Erika (37:52) She was like, yeah, like this whole thing, looks like it was generated. I was like, no, I actually wrote that, but I put a bunch cheese all over the place.

Brittany Ellich (37:59) no.

Bethany (38:01) Okay, I agree with that. think emojis help convey sentiment in written work. But I think you have to use just the most chaotic emojis possible. And then that helps. it’s like, clearly AI’s not reacting with LOL sob all the time.

Erika (38:11) Thank

Brittany Ellich (38:17) Valid.

Erika (38:18) Maybe

that needs to be our next prediction is what emoji is going to take off in 2026. What’s going to be the chaos emoji? All right. We’ll put a pin in that for next time. have time to think about it.

Alright, well for today, thank you everyone so much for tuning in to Overcommitted. If you like what you hear, please do follow, subscribe, or do whatever it is you like to do on the podcast app of your choice. Check us out on Blue Sky and share with your friends. Until next week, goodbye.