Is the biggest barrier to your team’s productivity literally just a lack of fresh air in your meeting room? This week on the Friday Deploy, Ben and Andrew dive into the rise of highly capable open source models like GLM 5.2 and the messy reality of running local AI for coding tasks. The hosts also discuss the cultural shift away from deep reading in a world obsessed with AI summaries, emphasizing the importance of protecting your first brain. Finally, they review a legendary tale from Meta's engineering history.
Show Notes
- The bottleneck might be the air in the room
- GLM-5.2 is the step change for open agents
- Viability of local models for coding
- AI Erodes a Legacy of Reading
- I Shipped a Facebook Feature So Fast Sheryl Sandberg Called an Emergency Meeting to Stop Me
Transcript
(Disclaimer: may contain unintentionally confusing, inaccurate and/or amusing transcription errors)
[00:00:00] Ben Lloyd Pearson: You know, Andrew, I feel like week after week on this show, we, we're just, we spend so much time thinking about, like, all the bottlenecks that are being created, like, with, with AI and, you know, all the processes we have at different companies and the tech stacks we have. But, you know, s- uh, after reading some of the articles we had for this week, um, I was, came to the realization that we don't always talk about the biological bottlenecks. Um, so I don't know, Andrew, after, after this article we, we read this week on how the bottleneck for your team might be the air in your room, and specifically how much CO2 is in the air, uh, are you walking around with a CO2 monitor now? 'Cause I'm kinda thinking about doing it.
[00:00:39] Andrew Zigler: No, I've kind of, I've kind of wanted to put one of those on my home assistant, I won't lie. But I loved reading this article about sometimes the biggest enemy in a meeting is the fact that you've been locked in a closed room with a whole bunch of people with poor airflow. I, as someone who has, uh, unfortunately suffered through office work before, um, I think this is a totally real [00:01:00] phenomenon.
[00:01:00] Andrew Zigler: You know, air quality in the workplace is a big deal. But, uh, it also points to, like, everyone needs different levels of air quality as well for their health, and it's good to be in an environment, like in my case, working from home, where if I wanna monitor my CO2 levels, I could maybe install something easily.
[00:01:16] Andrew Zigler: And frankly, there's less people breathing the air with me right now in my living room
[00:01:21] Ben Lloyd Pearson: Yeah. Yeah, and we'll include a link to this article in the show notes. But, you know, the TLDR is that there's some new research out that says that, you know, if CO2... It actually doesn't take much for CO2 levels to rise to a level that negatively impacts your cognition. So if you're, if you're sitting in a long meeting, uh, in a room with other people, or even if you're sitting in your home office just working by yourself, it's actually not that uncommon for you to be in an environment where, uh, the CO2 is actually causing you to have brain fog or to just make poor decisions.
[00:01:51] Ben Lloyd Pearson: So yeah, it was, it was a really great read. Something to be conscious of, I think, and, you know, just add it to the list of reasons that I, I need to go out and buy one of these CO2 [00:02:00] monitors, 'cause I, you know, I've thought about this a lot of times over the years and, and I think it would actually be nice to get some, some data on it.
[00:02:07] Andrew Zigler: Okay, well I'm gonna hold you to it
[00:02:09] Ben Lloyd Pearson: Yeah. All right. Well, maybe next episode you'll see one behind me. But yeah, welcome to the Friday Deploy, brought to you by LinearB. I'm your host, Ben Lloyd Pearson
[00:02:19] Andrew Zigler: And I'm your host, Andrew Zigler
[00:02:21] Ben Lloyd Pearson: And this week, we are covering the-- some open source AI models, the limits of local AI, the weekend feature that spooked Facebook's C-suite, and when AI erodes deep reading. Uh, Adam, uh, it... me, uh, not Adam. Adam's wa- is helping us record this. Andrew, let's talk about these open source models 'cause, uh, this is-- I'm excited about this actually, like what we're seeing here.
[00:02:46] Ben Lloyd Pearson: So what, what do we have with, uh, GLM 5.2?
[00:02:49] Andrew Zigler: Okay, GLM 5.2 open source model recently hit the scene under an MIT license, and, uh, many folks that have been trying out the different open source models that have been coming out in the [00:03:00] last few months, there's been a number of them, have started to circle around this one as, "Oh, yes, this one feels right in the coding harness."
[00:03:07] Andrew Zigler: And it's the first of an open weight type of its kind to be performing at the levels that we're seeing on benchmarks against foundation models. But also, too, it just has a healthy respect for balancing thinking, action, and tool calls, uh, in a way that other open source models have up until now kind of been experimenting with getting right.
[00:03:26] Andrew Zigler: And this also points to just the, um, significant foothold that open source models have gained in the last few months. We've been seeing a number of really huge high-profile releases that are competing with, uh, very large and expensive, uh, foundation models that, uh, we see as, like, the leading, um, ones on the market.
[00:03:45] Andrew Zigler: And so it's really interesting when on these targeted benchmarks you start to see these smaller, uh, open source players, uh, really start to rank up
[00:03:53] Ben Lloyd Pearson: Yeah. Y- you know, Andrew, you and I go back and forth on this, like, I told you so debate around AI costs. Like, [00:04:00] Fable comes back out last, last week and, and you win because AI costs are just getting higher and more expensive than ever. Uh, but I think this article a- and this development is actually a win for me. Uh, you know, it's proving that there is still that, like, downward pressure on AI costs. So I don't know, Andrew, what do you think? Do I deserve a victory lap on this one?
[00:04:19] Andrew Zigler: You do deserve a bit of a victory lap. I will say that I am not happy that I'm correct that API co- AI token costs have gone up. But the reality is, is that model routing's only going to get more important. You're right, Fable coming back on the scene just reintroduces a huge, uh, gradient of intelligence that you can be working with on the tasks, and the importance of choosing the right model at the right cost for the task at hand.
[00:04:42] Andrew Zigler: I think that becomes the new, uh, levers that we're all playing with. And, you know, we've been experimenting with how to do this, uh, for a while on the show. Been talking about open source models being a key component of how even huge companies are transforming their internal workflows and reducing their [00:05:00] bill, uh, which is allowing them to reinvest into their engineering.
[00:05:04] Ben Lloyd Pearson: And, you know, one thing that, that I, I think was really important that was pointed out in this article was how many of these low-cost and open source models are about six to nine months behind the frontier models in terms of the, the capabilities that they're developing, which kind of sounds like eternity in, in this day and age, but I think this is-- that's actually r- really what makes this so exciting because you remember back to where we were at the end of last year, you know, about seven, eight months ago, you know, we were covering the emergence of like the Ralph Wiggum loop and beads and how these new technologies were changing the nature of agentic coding. And that was the point in time where, you know, the frontier models got good enough for high degrees of autonomous coding, uh, and that practice started to go mainstream as a result of it. So now we're reaching that same point with open source models, and the options that we have for high capability [00:06:00] models is now exploding while the costs are also declining as well.
[00:06:04] Ben Lloyd Pearson: And, you know, as you mentioned, model routing, I mean, it's really the thing that we're missing. I, I feel like it's the biggest gap in terms of agentic infrastructure right now, is having some sort of layer that automatically knows which model is best suited for the task. Because like today, I feel, I feel like everyone is just sort of taking their own approach.
[00:06:21] Ben Lloyd Pearson: Like they're building these like homegrown custom solutions for it that are even like sometimes all the way down to like the use case level. It really feels like we're in like the experimentation phase of those mo-model routers, um, in a way that reminds me of where we were about a year ago with just agentic coding, uh, in general. you know, and we're seeing how like models like Fable, you know, we brought that up already, um, they're getting really good at delegating tasks to sub-agents, but we also need these services to choose the model for all of those subcha- sub-agents as well
[00:06:56] Andrew Zigler: Mm-hmm.
[00:06:57] Ben Lloyd Pearson: the cost whenever possible. And, and, you know, and I don't know that [00:07:00] Anthropic is really like incentivized to do that, right?
[00:07:02] Ben Lloyd Pearson: Because they, they kind of want you to, to use more of Fable because it makes them a lot of money. But, you know, we've been covering... And, and, yeah, and this, this model routing, I think that is the next big unlock within the agentic coding space. And we've been covering, you know, tokenmaxxing a lot here at DI. We, we just had a workshop at LinearB on this topic and how to be efficient with your AI usage. Um, and I f- I feel like, you know, AI token costs really are top of mind for every engineering organization who's rolled out new AI tools over the past year or two. And tokenmaxxing, it really kind of stinks when your costs are spiraling out of control, but it can actually be kind of an amazing force if you do it while minimizing costs and ensure that, you know, that velocity is actually translating to actual productivity gains.
[00:07:49] Andrew Zigler: Yeah
[00:07:50] Ben Lloyd Pearson: and, you know, this is stuff we talk to engineering leaders every single week about, which we, we really love. But yeah, you'll have to tell me what your agents think about GLM 5.2 once you've had the chance [00:08:00] to, to dissect it in your lab. Uh, so let me know how that goes. And in the meantime, I'm expecting this pressure from low-cost models to continue building for companies like Anthropic and OpenAI, you know, particularly how quickly the- their advancements can be replicated in these other models.
[00:08:15] Andrew Zigler: Completely agree about the pressure, and folks are gonna be pressured and, and needed to experiment on this. And I have already poked around on the model. It's pretty capable, and, I think that it's gonna have some really promising stuff for us in the future.
[00:08:28] Andrew Zigler: And one other thing I wanna point out as well, just in defense of Anthropic a moment ago, you said about obviously they want to incentivize us to use more of Fable and the, the expensive model, and that's totally true, but Fable is actually a first in class of its kind that has this kind of model routing built into the intelligence because it will reach out to Opus for some kinds of, uh, security flagged coding requests.
[00:08:52] Andrew Zigler: It will go out to Codus instead of doing it itself to avoid, um, you know, potentially it being used in, like, an incorrect manner. And [00:09:00] in the same way, it will delegate writing things out to Sonnet-05. So, um, it does think of itself as like, "I have these agents that I reach out to." Uh, Anthropic's been making a number of, uh, strives towards this because ultimately, uh, routing is a matter of intelligence itself.
[00:09:16] Andrew Zigler: Uh, you can't fully deterministically route everything. You need an intelligence to route the intelligence, and so it becomes a question of like, you know, who routes the router? And it turns out that maybe that's gonna be, like, the Anthropics of the world.
[00:09:30] Ben Lloyd Pearson: Just routers all the way down. And speaking of routers all the way down, let's talk about the viability of local models for coding as a, as a model to delegate your sub-agents to. Uh, so this article comes from friend of show, Birgitta Boeckeler at Thoughtworks. Uh, she spent about four weeks testing some locally run AI models for coding tasks and, you know, found that while the experience is, has improved pretty significantly in recent months, it's still too messy for most [00:10:00] developers to adopt without some serious effort. She did a whole bunch of scenarios and, you know, tested performance and, and model output. She pointed out that RAM is a, is one of the core constraints that, that all of this faces right now. Um, but she also tested out tool calling, you know, those sorts of things.
[00:10:19] Ben Lloyd Pearson: And there were some surprising findings in, in this that, um, I'll get into in a moment. But, you know, I always-- It's always wonderful to read what Birgitta has been up to. She has a very, like, a very, like, just robust way of analyzing th-the frontier of agentic coding, she really is affirming what I just said about us being in this experimentation phase of local sub-agent delegation.
[00:10:43] Ben Lloyd Pearson: So before I get into some of the things that stood out, I'm, I'm curious what you think about this, Andrew.
[00:10:47] Andrew Zigler: Yeah, we love exploring the ThoughtWorks memos here on the show. We've covered a number of them in the past year with these really great breakthroughs. I, and like you, really love this lexicon that Birgitta gives us for understanding the [00:11:00] different, components of what makes a good local coding model and the experience of using it, and it's a very complex challenge.
[00:11:07] Andrew Zigler: Folks often just stop at the model, "What model am I going to use?" But there's a huge amount of decision-making that has to go into allowing it to operate efficiently, and that's actually something we take for granted a lot when we use off-the-shelf harnesses like Claude Code that have so much of this baked in.
[00:11:24] Andrew Zigler: When you start to roll your own, you have to think about that piping yourself, and so Birgitta does a really great job of giving us a roadmap of what are gonna be the fastest constraints that you hit. RAM is absolutely the right thing to call out. You know, just a moment ago we talked about GLM 5.2. One of my biggest barriers in experimenting with it is just in the RAM that I would want available to it, right?
[00:11:45] Andrew Zigler: And so, um, you definitely have to think about, uh, all of these different constraints when you start to build with these tools.
[00:11:51] Ben Lloyd Pearson: Yeah, and there, and there was one finding in particular that really stood out to me, and that was around reasoning. You know, a lot of people, they like to, they like to turn on all the [00:12:00] capabilities with their models and get them to do everything because, uh, it, they kind of think, "Oh, that, that surely must make it smarter and able to make better decisions." But, uh, actually, the reverse may be true in some situations. So, um, if you have a reasoning model operating in a situation where, uh, the decisions aren't as clear as they need to be, um, it can actually get stuck in logic loops that will consume tokens and degrade the overall performance. So, you know, there's a lot of just, like, nitty-gritty details like that, where if you're someone out there who's listening to this and you're, like, ahead of the curve on experimenting with different models because maybe your job is to, to find the best model for, for your, your product that you're building or for your internal processes, um, there's a lot of really great insights in this article, uh, that, that can help you, like, understand how you need to approach the challenge. And there's also, like a, like, I think, a very clear warning a-about this, and that is that she really struggled to adequately measure and compare the [00:13:00] success of most models beyond, like, basic performance data, you know, things like time to response. So, you know, I, I'm gonna keep beating this drum for a bit.
[00:13:08] Ben Lloyd Pearson: Like, sub-agent delegation is this next big technical leap that we all need to go through, and I, uh, I love this article so much because I think it really gets into the details of why, why it is still such a big challenge using local models as sort of the lens for it.
[00:13:22] Andrew Zigler: Yeah. The reality is, is exactly that, Ben, that oftentimes it can be so hard to compare or use them in a daily fashion once you do get them set up for the exact reasons that you said. We get so used to, like I said a moment ago, about, you know, using Claude Code and having these high capability foundation models and these super souped up like Corsair style coding harnesses, right?
[00:13:43] Andrew Zigler: But when you start rolling your own and bringing your own components to play, uh, simple things like just having too many tools available, having too much thought, um, constraint on the actual model inference can just completely blow out its context window and just leave you [00:14:00] spinning when you're trying to use it.
[00:14:01] Andrew Zigler: Um, not to mention all of the finickiness of how different models will emit their tool calls and their thinking and the different ways that your harnesses have to handle it. And I haven't even begun to scratch the surface of prompt caching, token caching, which is really the underlying mechanic that keeps your Anthropic and OpenAI bill from being, uh, 100X what it is, and you need that too.
[00:14:24] Andrew Zigler: And so, uh, it only gets more complex. Be sure to check out Birgitta's wall of death that she gets from the LLM at one point. I cracked up at that because, uh, she follows this with a second memo, and it's at one point, you know, the local model just completely starts filling the screen with the same word over and over again.
[00:14:40] Andrew Zigler: And if you used, uh, ChatGPT, like 2.5, you... That r- that experience is very understandable and relatable
[00:14:47] Ben Lloyd Pearson: A- amazing. All right. Let's talk about how AI erodes a legacy of reading. Now, Andrew, I have to admit, I w- I just barely skimmed this article, so you're gonna have to tell me what it's about.
[00:14:57] Andrew Zigler: Oh no, I only barely [00:15:00] skimmed it too. What's this saying about us, Ben? I love that this article's just gonna hit it right on
[00:15:05] Ben Lloyd Pearson: read it.
[00:15:06] Andrew Zigler: It's gonna, it's g-
[00:15:07] Ben Lloyd Pearson: to
[00:15:07] Andrew Zigler: oh, okay, okay, okay. Well, maybe I'm just joking too, but why don't you take the lead?
[00:15:12] Ben Lloyd Pearson: Yeah. So this article is about the author, arguing that AI-generated content and AI-augmented content sort of accelerated the, the death of like deep word for word reading. Like when was the last time you had text put in front of you that you really considered every single word in that text? Um, and it's-- But the, the argument really is not that it's not just that there's more text than ever. Um, we've also had this phenomenon of, of like an eroded trust in authorship that makes people less willing to invest attention in any given piece, because the, the moment they think that maybe AI generated it, well, they, they think, "Well, the author didn't spend that much effort writing this, so why should I spend my effort, um, reading it?" And I think one of the big takeaways for here, [00:16:00] you know, and this extends beyond software engineering, it's really more focused at, on just all knowledge workers as a whole. You know, as content volume becomes unmanageable and increases over time, summarization tools feel necessary to stay current and on top of things. Uh, but the author wants to warn us that this creates a feedback loop that further degrades the quality of engagement with ideas. There's a lot of nostalgia, some contrarian opinions that, liked throughout all, all of this. Um, he also acknowledges that skimming is a very rational behavior given the volume that we're all dealing with today, um, especially if you're trying to make like value-based decisions, you know. but I think it's worth discussing for, you know, for all knowledge workers, but for engineering leaders as well. You know, teams are already navigating this trade-off with, internal docs proliferating, spec driven development, all these async communications. it can be healthy to question from time to time whether AI summaries help [00:17:00] teams move faster or are just quietly degrading like a shared understanding.
[00:17:03] Ben Lloyd Pearson: So, so Andrew, what did you pick up from skimming it or reading it end to end? I don't know.
[00:17:08] Andrew Zigler: There are a few key points of this that really resonated with me, and one of them was the idea that you have to have a summarization system these days to even keep up and to be reading things, and I couldn't agree with that more. I unfortunately have created all of these huge nets, especially in, like, the news world, that catch things and percolate them down, and I have things that matter to me that look for stuff.
[00:17:31] Andrew Zigler: And so this method, method of gathering all of the noise in the world and compacting it and summarizing it down, uh, is, is something that's been really important for me to keep up with. But then with that comes the temptation to only skim, to only read the digest, to only kind of skim across the surface, and this is what does cause the collapse of the conversation that this rightly called out, that, you know, then the folks that are putting out the content aren't getting the engagement or aren't getting the, uh, the [00:18:00] critiques and, like, the, uh, head-to-head conversations and the rebuttals and stuff.
[00:18:05] Andrew Zigler: And, you know, maybe this was a, a, a trend o- of older times where people would have their blogs, and they'd be writing, like, reply blogs to each other and, uh, arguing about, like, all sorts of foundational things, whereas now we just typically, uh, have our own silos where everyone's just throwing things out into the world, and you pick out what, what is interesting to you.
[00:18:25] Andrew Zigler: Uh, another part of this that was really, um fascinating was the idea of author trust and understanding where, uh, the content that you're reading even comes from. And this goes hand in hand with what you were saying about like summaries and, things like, uh, AI generated notes and stuff. Are they just piling things on top?
[00:18:44] Andrew Zigler: Are they actually, uh, helping us understand the o- the ownership, the authorship of like what we're trying to do here together? That's like a big question I think that comes into play with everything we read now. Um, as somebody who studied Latin and Greek in college, [00:19:00] I spent a lot of time reading really old stuff where the authorship and the even like the accuracy of the letters in front of you are questions and debated, uh, you know, across time.
[00:19:10] Andrew Zigler: And so you always have to approach what you're reading with this lens of where did it come from? How much was this translated? What was the legacy of it? How much of it even survived? And that I think the, the historicity of understanding where this stuff came from was fascinating. And we're seeing the same stuff play out now with AI generated content, I think especially when you get them feeding into each other and they're making noise and, and, and then maybe nobody's paying, uh, a- attention.
[00:19:38] Andrew Zigler: But I will say like one tactic that you can use you mentioned beginning like when's the last time you read something and really stared at every sentence is, you know, to slow down. I think that's the biggest recommendation I can make. It's okay to not read everything. You, if you create these systems that can capture and can digest things that matter to you, then that buys you the privilege to slow [00:20:00] down and to read, uh, because the things that you do wanna go learn and find out can always just go be a hop, skip away
[00:20:06] Ben Lloyd Pearson: Yeah. And I, and I completely understand the author's perspective in this article, but I, I think really what they're describing is an ef- effect that has been at play basically since humanity invented words to begin with. You know, originally only a very small group of p- people had the ability to read and write, and they were effectively the gatekeepers of all information. Very few words at th- at that time were created, and each word proportionately had a much bigger impact on, on people. you know, as literacy has become normalized, more words are created, and eventually we had the printing press, which just ratcheted up a whole order of magnitude. Um, and now we have effectively an automated or an autonomous printing press with all of these AI tools.
[00:20:52] Ben Lloyd Pearson: So again, generating words is easier than ever, and naturally word generation has increased yet another order of magnitude. [00:21:00] but then along with this comes like this proliferation of knowledge. You know, it's one of the big benefits of, of, of having language and words. you know, back when you only had access to a few books, it was, you know, and we're talking hundreds of years ago, it was really easy to memorize everything that was inside the books that you had access to because you had proportion-- again, proportionally more time to dedicate to studying the words in those books. But, you know, and as we generate more knowledge, it really is impossible for humans to keep up with all of it. So naturally, you know, just like AI has to compact its knowledge, we have to compact our memories and our knowledge. and, know, with that said, you know, we shouldn't always operate that way.
[00:21:40] Ben Lloyd Pearson: Sometimes you have to for, for velocity or, or other reasons. Um, but I do think there is a massive payoff to from time to time just being very intentional and spending that time to go deep on something and build your first brain, so to speak, rather than focusing on recycling knowledge into your second brain, into the AI [00:22:00] systems that you have. Um, and yeah, I, I was joking. I did actually read this end to end every word, and I slowed down to read it a bit to respect the point of the article. And, uh, yeah, and
[00:22:13] Andrew Zigler: Good job, Ben
[00:22:15] Ben Lloyd Pearson: yeah. So yeah, sometimes don't outsource all your thought to AI. I guess that's the, the point I wanna make. I, I think it's really critical to keeping a sharp mind All right.
[00:22:23] Ben Lloyd Pearson: I wanted to cover this last article about Facebook because, you know, I, I feel like it's, it's been so, so easy for us to throw shade at Meta. It is from time to time fun to also just look at some of the cool things that have happened there over the years. Uh, and this story comes from an engineer that worked there. Uh, it's titled "I Shipped a Facebook Feature So Fast That Sheryl Sandberg Called an Emergency Meeting to Stop Me." And this comes from, again, from a former Facebook engineer, uh, where they've recounted... It, it is a wonderfully written article where they recounted shipping a file upload feature for Facebook Groups over a single weekend. Uh, and this [00:23:00] resulted in basically an emergency meeting that involved Mark Zuckerberg, Sheryl Sandberg, and the CTO of the company. You know, basically three very expensive executives all showing up on this engineer's calendar out of the blue. Uh, and you know, the, the article really is just about how, uh, you know, Meta it ha- has historically had a, a, an aura about it of having a really good engineering excellence culture.
[00:23:26] Ben Lloyd Pearson: You know, giving a lot of autonomy to their teams, um, and letting them just take big shots at, at building new capabilities. And it was really just the, you know, the culture that, that fed into that. But yeah, it was a great, great story of how everyone thought it was gonna take months. The engineer produced it in days, and suddenly everyone's panicking about how quickly they can produce it.
[00:23:48] Ben Lloyd Pearson: But, you know, again, I, I, r- I, I don't always wanna throw shade at, at Meta, so it's nice to have an article that kind of harkens back to some of the great engineering culture that it has fostered over the years. What'd you think about it, [00:24:00] Andrew?
[00:24:00] Andrew Zigler: You know, uh, this kind of, um, story about suddenly on your calendar a whole bunch of executives dropping a meeting, and you're just a, you're just an engineer somewhere writing your code and shipping features for people, you know, that's definitely enough to scare the daylights out of you. Um, and so this i- idea that the article explores about, you know, there's this great deal of ownership at Meta about what, the features that they did build and what they shipped and, and how this did lead, lead to so many breakthrough, uh, pieces of technology, which, you know, undeniably we can credit Meta and their engineers to, to bringing into the world.
[00:24:33] Andrew Zigler: And so that kind of engineering culture is definitely something people strive for, but there is an element of like a survivorship bias to this story that I definitely pick up because I think in many cases, this is actually a reminder that as you do get more abstracted from these in the weeds technical shipping, delivering, write that 10,000 codes, do it in eight hours kind of deal, you need to, one, maintain really deep awareness of what your [00:25:00] agents are doing, but then you need to still be able to have the ownership of understanding what they're putting out the door.
[00:25:06] Andrew Zigler: Because ultimately, if you have a bunch of agents running wild and writing codes and shipping things and knocking out features like this, a story like this would play out today, and you-- maybe that feature would be, uh, have skipped around a security, uh, requirement, or it wouldn't have used this very specific kind of a library, right?
[00:25:23] Andrew Zigler: There's all sorts of nuance that we have to be responsible for as engineers to put into these systems. Um, and so ultimately, this is a reminder to not short-circuit the systems in place within your engineering org to get code out the door. I think as engineering, as an engineer, if you're hitting a constant barrier with your agents and trying to get things delivered, then that requires so- that, that then should be something we all bring attention to as a team.
[00:25:49] Andrew Zigler: Why is this now an obstacle? Because there are new bottlenecks places, uh, and we're constantly hunting down those bottlenecks now as engineering leaders, so
[00:25:57] Ben Lloyd Pearson: Yeah, well said. So Andrew, what have your [00:26:00] agents been up to this week?
[00:26:01] Andrew Zigler: Okay. Well, besides the, the digest and the reading thing, and then getting to hang out with Fable again, which was an unexpected surprise, I am now creeping up on the very end of my usage window. But some amazing things that, uh, we managed to get out the door this week, one of them was a web app for my phone I can have on the home screen of my phone that actually lets me peer into all of my beads, um, that I use to do all my tracking.
[00:26:24] Andrew Zigler: I can also see if any of my sessions are stuck because I've kind of fallen really deep into the loop engineering world. So I have a bunch of loops that are pushing stuff along all the time, and I typically just pop in there to see if any of them have gotten stuck. What about you? What about your agents?
[00:26:38] Ben Lloyd Pearson: Yeah, I feel like the mobile development is, is very quickly just becoming a cultural norm, I think.
[00:26:44] Andrew Zigler: I've had a lot of fun. I read this recent article from Simon Willison about how he just kind of creates random iOS widgets for things now because it's just so easy to write like a Swift app that's native to your Mac. And I was like, "Wait, that's genius." So now I just have like a bunch of, [00:27:00] uh, SwiftUI kit things that just pop out every once in a while, and then I throw them away
[00:27:05] Ben Lloyd Pearson: Yeah. Well, uh, yeah, I, I mean, our, our listeners may know that I, I was out last week t-touching grass, which unfortunately grass does not have the latest Fable models installed upon it yet, you know, so I can confirm that. Um, but since I've been back, I've, I've actually been focusing more on my first brain this week. So, you know, I've got some big projects that are getting kicked off, and it's, it's like I really wanna develop the ideas very, like get them really robust before I go kick off a whole bunch of agents to spin off a whole bunch of work from it. So, uh, yeah, it's, you know, I am kind of taking the, the article we covered to heart a little bit this week and being s- being a little slower, being intentional in building that first brain so that my second brain can operate a little better. All right. Well, that's it for this week's episode. If you enjoyed it, remember that everything we discuss comes back to one major challenge, and that is that engineering teams everywhere are wrestling with AI right now. Wrestling, I tell you. [00:28:00] It's writing code faster than ever, and your SDLC probably is struggling to keep up, 'cause I know ours often is. LinearB is the engineering productivity platform that shows you exactly where AI speeds up delivery and where it stalls, and we automate the bottlenecks so that your team can ship faster with total confidence. See how LinearB can help your engineering organization by checking out linearb.io. Thank you for sticking all the way to the end of this show.
[00:28:25] Ben Lloyd Pearson: Uh, if you're still listening, engage with us somewhere. You know, give us a thumbs up, give us a like, uh, leave a comment wherever you're watching this, or just reach out to us on social media. We always love to interact with the community, and of course, that, that all helps us grow the show. So thanks for listening, and we'll see you next week
[00:28:42] Andrew Zigler: See you next time
