Did AI agents just DDoS GitHub? Andrew and Ben are joined by Warp Founder and CEO Zach Lloyd to discuss the massive strain agentic workflows are putting on our infrastructure and why the "Monday Morning Commit Spike" is the new normal. They also dive into Steve Yegge’s reflective piece on the "AI Vampire" and the economic pressure on developers to output 10x results without 10x pay. Finally, Zach unveils "Oz," Warp's new platform designed to move agents off your laptop and into the cloud for better orchestration, security, and team collaboration.
Show Notes
- GitHub: GitHub appears to be struggling with measly three nines availability
- The AI Vampire: Read Steve Yegge's Article
- Open Coding Agents: Tim Dettmers/AI2 Research on Open Coding Agents (SERA)
- Listen to: The future of the terminal is not a terminal
- Warp Oz: Introducing Oz: The Orchestration Platform for Cloud Agents
- Blog: Introducing Oz: the orchestration platform for cloud agents
- Connect with Zach Lloyd: LinkedIn | X (Twitter)
Transcript
(Disclaimer: may contain unintentionally confusing, inaccurate and/or amusing transcription errors)
[00:00:05] Andrew Zigler: Welcome to Dev Interrupted. I'm your host, Andrew Zigler.
[00:00:08] Ben Lloyd Pearson: And I'm your host Ben Lloyd Pearson.
[00:00:11] Andrew Zigler: and joining us for this week's news is a good friend of the podcast, Zach Lloyd, the founder and CEO of Warp. And Zach, it's really great to have you back. For our listeners, if you haven't checked out Zach's episode from last year, we strongly encourage you to go back and give it a listen.
[00:00:27] Zach Lloyd: Thanks for having me back on. I'm glad, glad, I glad I earned a repeat invite.
[00:00:32] Andrew Zigler: Oh, of course. I mean, this'll be first of many more, I'm sure. Uh, 'cause we know, we knew back when you came on, uh, the pod and we talked about warp, which for those who aren't familiar is the agentic terminal. Like it was a really net new thing for me. And we talked for about everything under the sun, about a.
[00:00:48] Andrew Zigler: Developers and why they would adopt it. And today we have a whole bunch of news in store, uh, about and developments in our industry around orchestration, things that we've been covering week after week here on Dev Interrupted, [00:01:00] because age Agentic orchestration is coming in for everything in engineering right now.
[00:01:05] Andrew Zigler: And the greatest thing is that right now, you know, warp, you've just released a, a new product on this as well called Oz, and we're gonna talk about that a little later in our news Roundup. First, we do have to cover some of the highs and lows of this week in tech and the things that we all lived through as we evolve into the agent developers of tomorrow.
[00:01:25] Andrew Zigler: And the first thing we're gonna talk about is GitHub maybe struggling to keep up this week because if you were like me or many other developers who have an orchestrator at this point, you might have rolled out of the weekend doing about a thousand commits. So. When it came Monday morning and you and all your coworkers were turning on these token machines, something dramatic very happened.
[00:01:46] Andrew Zigler: It very much happened to GitHub and uh, if you were like me, you probably, uh, already know that GitHub was down. So Ben, you know, what did you think of one of the, one of the largest code forges in the world, having this tumultuous start on Monday, and [00:02:00] what does it mean for us?
[00:02:01] Ben Lloyd Pearson: Yeah, I mean, it, it really points out how, like, as we become more dependent on ag agentic systems, like the infrastructure that runs all of this stuff is like more important than ever because like, like if Claude goes down, then there's like significant portions of my job at this point that I'm like basically incapable of doing.
[00:02:18] Ben Lloyd Pearson: Like I could do it, but like the manual effort it takes for me to replicate what Claude would've replicated for me. It is just like, like, it just doesn't make sense. Like I just wait for it to come back up, you know?
[00:02:29] Andrew Zigler: Right.
[00:02:30] Ben Lloyd Pearson: but yeah, it's, it's, uh, yeah, it's, it's pretty crazy. Just how, how impactful when you're, when you're using AI to accelerate everything, how, how big the gap feels when your infrastructure goes down.
[00:02:42] Andrew Zigler: Yeah. What about you, Zach? Did, did y'all feel it over there when GitHub wasn't responding? Monday
[00:02:46] Zach Lloyd: Yeah, this, this was the day before we were doing our biggest product launch of the year, and all of a sudden we couldn't, you know, we couldn't see dips, we couldn't make commitments. It was, uh, like horrendous timing [00:03:00] for us. And it was, we were just trying to find ways to work around it and like. We were coming up with creative things like we have a code review feature in Warp, so you can do it like locally.
[00:03:10] Zach Lloyd: And so we were just trying to find ways where you can bypass GitHub, but it was just horrendous timing for us. And I do think it's like, I don't know if you all know like what the root cause if, if they've talked about it at all, but the, I would assume that they are starting to strain because the amount of AI generated code and GitHub activity has to be going through the roof right now.
[00:03:31] Andrew Zigler: It is, it's skyrocketing. It's like code agents like Claude Code are committing about 4%, 5% of all commits on GitHub right now. And there's actually this chart that I saw on Monday, I'm sure we can include it in the show notes of the commits, uh, going up and up and up and getting steeper. And then right around the beginning of February it just skyrockets and suddenly you have slope on slope growth.
[00:03:55] Andrew Zigler: And I think that's a moment where a lot of folks started to really understand [00:04:00] orchestration and the power of running multiple agents at once and then then being able to scale to do so. So GitHub, I think just maybe wasn't ready for that kind of wave. I don't know if we've gotten an official kind of like breakdown on what happened but um maybe if you were like some, uh, folks too. Like at one point my agents recommended like, should we spin up our own git Forge? Like is GitHub gonna really be in the way of people? Uh,
[00:04:26] Ben Lloyd Pearson: Yeah, we had, I mean, we had Jeffrey Huntley on here, like basically just describing how he is building his own version of basically everything, because why not? So, I mean, why not build these backup plans at the very least that. Allow you to, to deal, to circumnavigate the issues, you know, but, but yeah, I have to imagine if, like, if your auto-scaling is built around like a mon a standard like Monday to Friday work week for like humans, like agents are like just nonstop working like over the weekends and, and you know, there's probably a pulse when everyone shows up Monday and like goes, tells their agents to do a bunch of work.
[00:04:59] Ben Lloyd Pearson: You know? I just [00:05:00] wonder how much that is impacting like GitHub right now. And probably other companies too.
[00:05:04] Zach Lloyd: Yeah, maybe they're vibe coding some of their infrastructure also, and that's leading to props. I, I don't, I,
[00:05:10] Andrew Zigler: Maybe it's eating itself.
[00:05:12] Zach Lloyd: it's, it is, I think, harder to build super reliable software with, with these coding agents right now. Maybe it's eating itself. Uh, I have heard chatter also of like, it time for a new piece of infrastructure at that layer of the stack that is more agent native?
[00:05:30] Zach Lloyd: I don't totally know what that, what that means, but it's definitely an interesting thing to think about because the sort of, all of the paradigms that we built for people to not necessarily translate perfect to agent first development. Like, have you guys thought about this? Like what, what, what would the evolution of this look like?
[00:05:47] Andrew Zigler: You know, I've thought about this too, about, there's a lot of things right now about how we code that we've put there as crutches for us as humans to be able to code. And with things like agents, it challenges us to, like, how much of that can we rip [00:06:00] away. Why does the agent have to write in a language that I can, that's human readable to me?
[00:06:04] Andrew Zigler: How can we get closer to a more machine and deterministic language? Why does, uh, why do I need an interactive shell in these kinds of ways? Like, what can the agent do with all these pipes and tubes in the background that can fundamentally change how data is, is used? I, I honestly think about how da uh, agents have transformed software and it's more like TCP or ip, right?
[00:06:28] Andrew Zigler: It's like now you can tokens in and you can pipe output and pipe execution somewhere else. And it's just a different way of building. I, I think a lot of layers will go away, get replaced. Um, I think it's interesting to study what will happen.
[00:06:42] Ben Lloyd Pearson: I think this might be a, a great way to take a dark segue into our next story on,
[00:06:47] Andrew Zigler: Oh yeah.
[00:06:48] Ben Lloyd Pearson: the AI Vampire, this latest article from, from Steve Yegge, someone that we're both great fans of Andrew, and I know this one really resonated with you. So, so walk us through what, what's going on with this article?
[00:06:59] Andrew Zigler: Yeah, so [00:07:00] we read everything Stevie Yegge reads, uh, or writes on this um, when we, when we cover it here, especially recently. And he, his most recent missive here is called the AI Vampire. And it's a reflective reflection of himself um, about his experience working with Gastown and the like, really rapid culture that has wrapped around it.
[00:07:18] Andrew Zigler: Steve is someone who's spoken verbally, you know, very much out before about his, uh, experiences with burnout at different companies. And so this is really his reflection on how Gastown accelerates developer work towards burnout and about how working in this unrealistic way perhaps that's unsustainable, is what he was reflecting upon and.
[00:07:36] Andrew Zigler: You know, first off, it, it was really interesting to get into his head about some of his own guilt, but interest around Gastown. There's so much conflict, I think, in with him in this article. But the biggest thing that stood out to me was like his own reflections on like what it means for everyone else.
[00:07:50] Andrew Zigler: Because like, you know, Steve is a really seasoned engineer. He is 30 plus years of engineering. Engineering experience at every level of the organization. And he is really reflecting on the realities of junior [00:08:00] engineers and mid-level engineers picking this up and then accelerating and what does it mean for them in their careers and their ability to make money within our economic system.
[00:08:08] Andrew Zigler: So really great article, A really interesting reflection. I is reads like someone who's been held up right against the fire. And I think there's a lot of wisdom in it. Ben, what did, uh, you think about it?
[00:08:19] Ben Lloyd Pearson: So, so yeah. And I, I think this ties in really well to the points that you brought up, Zach, on, on more agent forward tooling or, or tooling that is built for the agent space. You know, 'cause I really think we're in this like awkward transition period where. You have this like small group of, of people who have figured out how to like 10 x significant portions of their work.
[00:08:39] Ben Lloyd Pearson: You know, they're not maybe 10 X overall, but at times they are operating at that speed relative to where they used to be. But they're still surrounded by all of these like organizations, these processes, tooling and systems that weren't designed for like this scale of things. And I've, and I've been thinking a lot about how we're gonna continue to extend, uh, agent orchestrators, which is why I'm really glad we have what we'll get [00:09:00] to, warps launch here in a minute.
[00:09:01] Ben Lloyd Pearson: Uh, because I think it ties into this really well. But I, I have this sort of mental model that's starting to emerge where like, you know, everyone we've been covering in this space so far, like Yge, Geoffrey Huntley, Jeffrey Emanuel they've all built these like single purpose personal orchestrators that is like their own personal mental model for how like these orchestrators should work for them.
[00:09:23] Ben Lloyd Pearson: And, uh, you know, we're missing the layers that connect those personal orchestrators to other things within their team or, uh, organization. And I almost wonder if like the future is like layers of orchestration for this, where you have like the personal orchestrators that connect through team orchestrators that connect through like organization and company orchestrators.
[00:09:44] Ben Lloyd Pearson: Like this is like, I feel like this is thinking way off in the future, but with how fast things are moving, I, I really have no idea anymore. But yeah. So Zach, I'm curious like how you felt about this article as someone who's working really heavily in this space.
[00:09:57] Zach Lloyd: Yeah, I mean to, to the [00:10:00] orchestrator point first. So my, my thought on that is that we just need the right primitives. And then you can build like gastown. I don't know if you all have used it. It's cool. It's a very opinionated like. Orchestration system with like pole cats in the mirror and all
[00:10:16] Ben Lloyd Pearson: Yeah,
[00:10:16] Andrew Zigler: Oof. Yeah. We've covered, we covered the very, the very colorful metaphors here on
[00:10:20] Zach Lloyd: uh,
[00:10:21] Ben Lloyd Pearson: his own metaphors that are wonderful, so.
[00:10:24] Andrew Zigler: Yeah, the metaphors, I, it's been a rollercoaster.
[00:10:27] Zach Lloyd: So my feeling on that versus like, you wanna do like Ralph Wiggum or you wanna do cloud code teams is like, I don't really know. And I think there's a whole bunch of organizational systems that work for humans and I'm not convinced there's gonna be like a one size fits all thing for for agents.
[00:10:43] Zach Lloyd: But what I do believe and like what we've tried to do, like the, the future that I see is that you're gonna need primitives. And the primitives are like, you need agents that can kind of like run off your laptop. You need that, them, that, to be programmatic, like API driven, they need some way of like passing messages.
[00:10:59] Zach Lloyd: So there, [00:11:00] I just think there's all, what, what we wanna build right now is just like the primitives and let people organize these kind of like agent teams or agent organizations on top of them. So that's, that's the approach of warp. And then for the, like the, the AI vampire thing, the, the quote that stuck out to me from that was like, as an engineer, you know, if you, if you get really competent in using these coding agent tools, you can 10 x your development, but you don't get paid 10 times more for doing that.
[00:11:28] Zach Lloyd: Uh, it's like, and so it's, it's like all of a sudden we all expected you 10 times the work for like, you know, what we were doing before. And then he is like, well, or you could. Spend. Uh, and, and if you do 10 x the work and you don't get paid 10 times more, the company captures all that value. Whereas if you just you know, use the agent and you only work one hour a day, but you have your same output as before, then you capture all of the output for yourself.
[00:11:52] Zach Lloyd: And, and, but he also makes the point like, no company's gonna allow that. So it does. It's just like, it's like, what is [00:12:00] this? How does this change the expectations of engineers, I thought was an interesting thing. And like as someone who is constantly running these agents, I do feel the pressure to ship more.
[00:12:10] Zach Lloyd: Like even during this podcast, like, you know, I, I have an agent running, I've been working on this thing in the background where it's like, I want it to go all the
[00:12:18] Andrew Zigler: me too.
[00:12:19] Zach Lloyd: And so, you know, you, you feel like you're wasting time if you're not like, multi-threading these agents. And, uh, I think that's like a lot of pressure for an already kind of like, for engineers who are already under a lot of pressure.
[00:12:33] Zach Lloyd: Especially for like the junior engineers. 'cause like there's also this danger if you're, if you're early in your career and you're using these agents, it's very easy to like. Create a lot of like, fury around using them, but actually not have a lot of productivity gain from them. Like to like, like have them do a lot of stuff that can't actually be shipped.
[00:12:50] Zach Lloyd: And so I think I, I, I find that like one of the more frustrating things where it's like, I don't actually know if I'm like gaining from these by having them work all the time. I feel like it's straining [00:13:00] my attention span a bunch to like manage all of them. So I don't know. It's, it's a weird state that we're in right now.
[00:13:06] Zach Lloyd: It's also gonna change extremely quickly. We're not, it's not gonna be in this state for very long.
[00:13:09] Andrew Zigler: Yeah, it's gonna constantly be evolving and the way that, what you're touching on it, it covers what Steve said so well, but also too, it, it covers an article that we, we touched on recently about dark flow and about how like, sometimes vibe coding is like, or your age agent coding is like a slot machine where it's like you're just like putting in attention and tokens and hoping you get the output you want and then you're like putting all of these like extra coutre malt to try to get there.
[00:13:31] Andrew Zigler: And so it's, it was that, that article when, when we covered it, you know. I felt I had a bit of a pessimistic view because it ultimately does really revolve around the person using it and how they use it. And, and ultimately, like, like what you said, I, I feel the same pressure all the time to convert like the tokens available to me into like output and execution and like I have a lot of tokens available to me, so that's a lot of pressure.
[00:13:53] Andrew Zigler: Right. And I think a lot of engineers feel the same way.
[00:13:56] Zach Lloyd: Yeah.
[00:13:57] Andrew Zigler: So I wanna jump into our next, uh, story just [00:14:00] real quick before we get to Oz. And this next one is about, uh, leaning more on the research side of things. Uh, we're, we're in touch with AI2, as a research lab around um, AI and AI research. And he, they send us really cool developments from their lab all the time.
[00:14:13] Andrew Zigler: This most recent one is from Tim Dettmers talking about open coding agents. And how they were able to, uh, train them on top of specialized code bases to get really, really first in class model performance from open weight models that were then heavily, uh, trained on top of a target code base. And this is a really interesting development because as part of the research.
[00:14:36] Andrew Zigler: Their research team discovers that this problem ultimately broke down to like three or four critical failures, uh, that once they were able to address in like a systematic way, resulted in this open weight models exceeding the capabilities of like a foundation model teacher, like a Claude. A quad or a a codex.
[00:14:55] Andrew Zigler: This is interesting and the implications of this are powerful for people that are trying to use AI [00:15:00] that's tailored to their code bases. Right now there's a bit of a brownfield problem with AI and age agentic development, like what you're saying, Zach, of like, you know, did they make code that you could ship?
[00:15:10] Andrew Zigler: That's a really different question of like, did they make code that like, oh, they could use or like save some time for them. Like the stakes are so different there. And so the idea of being able to fine tune these agents that are from open weight models that are highly specialized on a very target code base um, I think that opens a lot of doors for how people can build on top of their own kinds of models.
[00:15:30] Ben Lloyd Pearson: Yeah, I think, uh, LLM or efficiency gains for LLM models are really like an underappreciated focus area right now. Like Zach, you brought up, uh, being able to run stuff like on your own hardware, I think is like it. A really, an unexplored area within, uh, a lot of, I mean, it's not totally unexplored, but it's not very matured yet.
[00:15:51] Ben Lloyd Pearson: you know, most of us t today we just kind of pick like, this is why we spend so much tokens. We, we just kind of pick whatever our favorite model is and we just send everything via [00:16:00] API over to that, you know? But I, I feel like a lot of, when, when we make this more efficient, a lot of the tasks will be able to.
[00:16:07] Ben Lloyd Pearson: Be handled by, uh, local models in particular. And yeah, this is, this, it's really detailed research into, to, to how to train agents to be more knowledgeable about private code bases where there aren't a lot of like general purpose lessons that you can apply to them. And that's, I I'm hearing frequently, this is a big problem for a lot of organizations, so it's definitely something that we need to solve.
[00:16:30] Zach Lloyd: Yeah, that, that was one thought I had when I read it. So, Warp is built on like a million lines of custom rust code, which I dunno if I would've done that decision again. Uh, it, it, there's, there's, there's good parts to it. It's tough though because the um, no, I think it's actually great from
[00:16:48] Andrew Zigler: That's too real. That's very real.
[00:16:50] Zach Lloyd: product quality perspective, but from a, like, uh, the agents don't know our UI framework and so they will often make mistakes 'cause they'll assume it [00:17:00] works like some other thing, like react or whatever. It doesn't work like that. It works in our own custom way. So from a like user of a model like this, I think it would, it definitely piqued my interest. And then from a founder in the coding agent space where we're, you know, we, we offer Claude and Codex and Gemini and some open source models in our app.
[00:17:20] Zach Lloyd: Uh, like anything that can create more competition or more options for our users there is great. So, you know, local LLMs. Awesome. I have a, a concern that like Claude and Codex are gonna be a sort of oligopoly where it's like. You, you, you know, people building on top of them don't have much choice. So I really want, I really want a bunch of choice there.
[00:17:44] Zach Lloyd: So I, I love developments like this.
[00:17:46] Andrew Zigler: Yeah, really well said. I, I'm totally agree with you there.
[00:17:50] Ben Lloyd Pearson: Yeah,
[00:17:51] Andrew Zigler: and you know, with.
[00:17:52] Ben Lloyd Pearson: we, we love covering the competition between frontier models. It's, it's, uh, very fun to watch how hard they're working for our attention.
[00:17:59] Zach Lloyd: yeah.[00:18:00]
[00:18:00] Andrew Zigler: their neck and neck right now inference right now costs what it costs, but one day it's gonna cost something very different and I imagine it'll be a lot more expensive. So I just am intrigued to see what happens with like competitors and the ability to even use those own models in your own machine is interesting Capability.
[00:18:16] Andrew Zigler: I think it's good to scale down while we're all scaling up is how I'll frame it. Because we're all trying to get to like those really big heights and you can't, if you're like lugging all of this baggage from yesterday.
[00:18:27] Zach Lloyd: The, the only way it'll get more expensive in my opinion is, is if there's like, uh, anti, an anti-competitive nature to it. Otherwise, at any given level of intelligence, like the actual cost per token goes down, it's only if certain companies have market power here that this will stay super expensive. Then my, my hope, or I think even my prediction is that, uh, for coding in particular, you're not gonna need to be at the frontier for that much longer in order to get good coding performance.
[00:18:55] Zach Lloyd: And so I think that will also open the market more, which is what [00:19:00] I, again, I'm, I'm very biased here, but like that is really what I want is like a market where people who are building this space are competing on the quality of the product, not the cost of the tokens, which I think is a little bit what's happening right now.
[00:19:14] Andrew Zigler: Absolutely. You know, I'm excited to see kind of how these things evolve because the things that we take for granted and we use every day, they're continuing to change and new things are coming into our, into our view that we can now finally see because of the things that we've been building yesterday and.
[00:19:30] Andrew Zigler: So I, I wanna get to the topic of the day, uh, which is your new release, Oz, the orchestration platform for cloud agents. And I just wanna open it up to you and maybe tell us a little bit about Oz, what it is and, and where the idea came from.
[00:19:45] Zach Lloyd: So, Oz, which we, we launched earlier this week is a, like you said, it's a platform for launching and orchestrating cloud agents. The, uh, sort of problem that it's trying to solve is, is getting agents off of individual developers [00:20:00] laptops, and the reason that's becoming a problem. Is, there's a few things.
[00:20:05] Zach Lloyd: So one, if you're someone who is now running like three or four agents locally, you'll start to find that you're gonna run out of CPU or memory or disc and it's gonna slow it on your computer and that you, you're gonna wanna multithread more. And so. Oz makes it very, very easy to, to do that from a more like, sort of like enterprise or business perspective.
[00:20:25] Zach Lloyd: What Oz is trying to do is make it easy for companies that want to really go all in on agents beyond just like giving individual developers agents as a developer tool, but like deploy agents across the whole company. If you want to do that, you want an easy way of getting those agents into the cloud.
[00:20:41] Zach Lloyd: So you want things like. Sandboxing, uh, you wanna be able to see what all the agents are doing as they're working. Like right now, there's no visibility. Every individual engineer is like just running these on their laptop. You wanna be able to secure them. You wanna be able to get an audit trail, and you want these agents to be able to integrate into your developers' workflows.
[00:20:58] Zach Lloyd: So Oz [00:21:00] is just trying to make that really easy and build the primitives kind of almost like Vercel super base, but for spinning up cloud agents. And so yeah, that's, that's what we, uh, that's what we launched this week.
[00:21:13] Andrew Zigler: That's what it makes me think of. It's like there's so much value in that being like. It for sale of where agents get deployed. And I think everything that you've addressed and like what we need I really feel that as somebody who's like does um, agentic orchestration to get a lot of my job done and I write a lot of code with it, like I did have to move to the cloud to support.
[00:21:32] Andrew Zigler: My throughput because they would bring my laptop to its knees. And honestly, it became a little like if someone walked by and saw my screen, it became like, concerning. So it just was better to move it all to a, uh, somewhere else. So now I literally do all of my coding, like through SSH, to like, you know, just something that's sitting out in the middle of America somewhere.
[00:21:53] Andrew Zigler: And I just hope there's no tornadoes later this year. And so honestly from there I'm thinking like, and that's the, and when I [00:22:00] set it up, I was like, and this will be the last time I ever do this, because either I'll use this long enough to where it'll build the next one for me and I'm not even gonna have to think about it.
[00:22:08] Andrew Zigler: Or someone else is going to figure out why I had to go and rent a VPS in order to get. This to work and they're gonna set this up in a way where I could do that. Right? And so I, I love this. Just wanna say from the beginning, 'cause I see all the value as someone who builds agents and I share them with coworkers constantly, the whole idea of like, okay, now I have to like, rewrite it and make it like relo it to get it
[00:22:32] Ben Lloyd Pearson: Package it.
[00:22:33] Andrew Zigler: state.
[00:22:33] Andrew Zigler: And oh, I need to do some serverless function on Vercel now. And I'm like, how do I even know what's working? Like there's, so I, uh, I really love the value of Oz. I'm really excited to check it out.
[00:22:44] Zach Lloyd: Cool. Yeah. The, the model that we have is less like you rent a Dev box in the cloud and it's much more like a, like a lambda model for agents. So just in the same way that locally in Warp, you might fire off an agent to do something or you might [00:23:00] fire off Claude code. Uh, you can just be like, okay, I wanna fire this off, but I want it to run in the cloud.
[00:23:05] Zach Lloyd: So that's like the simplest use case. But the other cool use cases that having these things in the cloud enables is like. It's more like automations. So for instance, like we have, uh, an agent that anytime we update our code, looks at our documentation and sees if it needs to be updated, or having an agent that writes our weekly change log, or we have an agent that's running pretty much constantly that's looking for patterns of fraud and abuse.
[00:23:30] Zach Lloyd: Uh, 'cause we, we have like a free AI tier to get people to try it. And so thinking in terms of automations and then even thinking in terms of like. If you're building apps, where can you put agents? And so, you know, we, we have an internal app that we built where I, uh, like I built this thing, it's like, lets you triage GitHub issues and what, you know, you, you can run an agent to de-dupe the issue.
[00:23:54] Zach Lloyd: You can run an agent to fix the issue. And that's all powered by by Oz. And so it's all [00:24:00] API driven and CLI driven. It's all like a program first approach to um, to launching these things.
[00:24:06] Andrew Zigler: I love it. It's addressing a major need. I feel it like Ben, Ben feels this too, 'cause
[00:24:11] Ben Lloyd Pearson: I mean, we were.
[00:24:11] Andrew Zigler: stuff that he wants to use and
[00:24:13] Ben Lloyd Pearson: Yeah, I mean, we were just talking about automatic change logs like this week, literally.
[00:24:17] Andrew Zigler: Yeah. Yeah.
[00:24:19] Zach Lloyd: it's super. So in the, the way that we approached this was through using skills. So like, you know, the, the skill standard, we basically, one simple way of thinking of automations is you can just put like a skill on a timer and run in the cloud. Simple as that. So a, you know, you just have to give it access to your, you know, build a docker environment for it.
[00:24:38] Zach Lloyd: And like, then you have a skill that's like automatically making updates to your change log and that kind of thing.
[00:24:45] Andrew Zigler: Yeah, skills are amazing. And so the idea that you're using these same basic ai, you know, agentic principles underneath to scale and build this foundation, it's, I, I think that's how this really, uh, the, the infrastructure that will stick around, [00:25:00] uh, will come into being because it acknowledges that like we have to build with these just new primitives, these new starting points, right?
[00:25:07] Andrew Zigler: Uh, and I, I think the, like, the really cool thing that stands out to me about Oz is the ability to distribute and share your gains from AI in a more healthy way. Going back to what we were talking about with like the do a hundred x or 10 x or output, and what you don't get paid 10 x or a hundred x more and then you burn out and then like, you know, is that fair to even your coworkers?
[00:25:29] Andrew Zigler: Like what's the value? System of that. Instead, this actually challenges and invites those folks that are getting the most of those benefits to find a way to distribute it more broadly to other people and other teams, because now there's no excuse for why you can't build that agentic thing that they need in finance for the last, you know, four months or whatever.
[00:25:48] Andrew Zigler: Right.
[00:25:49] Zach Lloyd: Correct. So the, yeah. One of the things that we, we did in Oz is that every. Every time an agent runs, no matter how you run it, if you run it through our CLI or through our [00:26:00] API, you could run it through our web app. It's, you can, it's shareable and it's like a team construct. It's not an individual construct and it's behind.
[00:26:07] Zach Lloyd: So you could be like, okay, I want these other engineers on my team to, to be able to sort of step in, see what this agent is doing. You could have multiple people in there at once actually, who are guiding and steering the agent, which is pretty cool. You get all of the, like, whatever the agent does, whatever it produces lives on in the cloud.
[00:26:24] Zach Lloyd: And so I think this is the basis of like an agent memory across an organization. Whereas right now, again, it's all just like local in your terminal session, which is not, it's impossible to build on if that's the primitive you have. Whereas if every agent conversation is, is a sort of cloud synced object, then I can do something and my coworker on the team can continue from that state.
[00:26:47] Zach Lloyd: And it's, it's like a pretty magical thing. So like I said, we're trying to build these primitives for what we imagine like. People and companies that are building real software are gonna want to be able to do this [00:27:00] at scale.
[00:27:01] Andrew Zigler: from your perspective, I'm sure you see even more places where this will go and you're like, oh, I can see what people would build on top of
[00:27:08] Zach Lloyd: Totally. So I was, I, we were just watching, uh, we just had our, our standup and saw a demo of, of what one of our like partners built. And it was, it was amazing 'cause he built this thing where, um. he's letting users of his app build the app, meaning like, when the user of the app is like, uh, I wish that the, like they could just submit a feature request in the app and, and then Oz builds it and they'll, and almost has the whole flow of users directly building the app that they're using, which I was just like, oh, that is such a cool creative use
[00:27:43] Andrew Zigler: That's so cool.
[00:27:45] Zach Lloyd: Yeah.
[00:27:46] Ben Lloyd Pearson: I, I've seen, I've seen some, some apps that are starting to do that, and it's, yeah, it is really profound to see the next iteration of this.
[00:27:52] Zach Lloyd: yeah, yeah. It's super fun.
[00:27:55] Ben Lloyd Pearson: Yeah, Asia orchestration is on like everyone's mind right now, so I feel like this is very [00:28:00] timely. I love new products coming out to support this type of stuff. We'll, we'll link to, in the show notes, to the article there's some really cool examples of what Oz has been used for.
[00:28:09] Ben Lloyd Pearson: So our readers or listeners definitely need to go check it out.
[00:28:13] Zach Lloyd: Cool.
[00:28:13] Ben Lloyd Pearson: yeah.
[00:28:14] Zach Lloyd: I.
[00:28:15] Andrew Zigler: Any other last words you wanna le uh, leave on that note about, about Oz and why we should go check it out.
[00:28:20] Zach Lloyd: no, I mean, just we, we, we would love feedback. I really wanna see people build cool stuff on it, like that was. That was
[00:28:27] Andrew Zigler: when's the, when's the hackathon? When's the
[00:28:29] Zach Lloyd: so we are, we are, we are doing, we are working on that actually. We're gonna do a hackathon. I'd love to see cool demos. if you start to get creative with what you can do once you have these sort of programmatic cloud agents, it's like you can do such cool shit with it.
[00:28:42] Zach Lloyd: So, uh, that's it, it's a it's at oz Dev or warp Dev slash oz. Either we'll get you there and you can try it out.
[00:28:49] Andrew Zigler: Amazing. Well, we're gonna share those links and I'm gonna be, I'm gonna back in your inbox about that hackathon.
[00:28:53] Ben Lloyd Pearson: Make sure you invite Andrew. He's gotta be there.
[00:28:57] Andrew Zigler: Amazing. Okay, great. Well, you know, uh, huge [00:29:00] thanks to Zach for joining us and giving us a first look at Oz and for joining us on our news journey this week, because it's been a pretty wild one. Uh, we all have lots of show notes, links where people can go and check out Zach and, uh, what he's building, uh, you know, Oz over there at Warp.
[00:29:13] Andrew Zigler: Um, As well as check out his episode here on Dev Interrupted, because remember he was a past guest here and his episode about Warp is really amazing. Uh, you can see the trajectory of how this stuff is evolving by listening to Zach then and Zach now. Um, And so definitely be sure to be tuning in. And remember, if you're only listening to me and Ben and our guests here on the podcast, then you're only getting half the story.
[00:29:34] Andrew Zigler: Uh, so be sure to subscribe to Dev Interrupted on Substack or on LinkedIn. Uh, we drop a full newsletter with each of these and it has a lot of links to, uh, further articles and things you can learn. So, uh, be sure to check it out and continue the conversation there. And um, thanks y'all for tuning in. Any exciting, uh, weekend plans on your ends?
[00:29:54] Zach Lloyd: I'm flirting with the idea of going skiing. We live near a ski area. We're supposed to get some snow this [00:30:00] weekend. I hope I can do it.
[00:30:01] Andrew Zigler: Ben loves to
[00:30:02] Ben Lloyd Pearson: I'm with you on that right now. I'm just trying to not be sick, so
[00:30:06] Zach Lloyd: Nice.
[00:30:07] Ben Lloyd Pearson: time of year.
[00:30:07] Zach Lloyd: on. Yeah,
[00:30:09] Andrew Zigler: Well, it's super sunny here in LA so there's no skiing, but I'll have to text some pics. And y'all have a good rest of your weekend.
[00:30:15] Zach Lloyd: You too. Thanks for having me.
