スタンフォード大学のAIコーディング講義はこう変わった

I made it into the Waymo with Mikael, Stanford lecturer, head of AI at Monaco. Here we go. What have you been doing with this Stanford course of yours that is very famous now? Yeah, it's been a fun time. So we're about, this is like, I think week five now of the Stanford course. Yeah. Yeah. Titled The Modern Software Developer. The Modern Software Developer. That's where it's at, yeah. And you have an email that goes out. I have an email. I do, I do, yeah. So there's a newsletter, themodernsoftware.dev. I have the domain, all that good stuff. But the course has been a ton of fun. I mean, this is the second year that I'm doing it. And compared to last year, there's just like a huge overhaul of the curriculum. pretty much none of i mean i literally i think 80 of the stuff i threw out over over the last like two months because it's just once opus four five came out december you know december slash november

december of last year when was the last year's course last year's fall so it's always in the fall and so the the last course uh in 2025 was september through like early december of 2025 and so for that like window of time nothing changed too much in terms of the model i was good i was good it was stable it was like relatively stable and then december rolls around and then uh you know things blow up in terms of capabilities people jump on the claudico bandwagon uh through like that's basically what everyone did over the over like december break you know during the christmas holidays and new years people came back early january and they were just there was a manic craze around a cloud code. But it was pretty justified. I mean, the models got so good that we could start doing some really interesting long-horizon Asian tasks. Anyway, so because of that, I had to rethink the entire curriculum. It's just like, well, what's now happened in the last nine months since cloud code became... So now you're talking about harnesses,

you're talking about loops. Harnesses, loops, software factories. We're talking about fully automated AI code review when not just, hey, add a GitHub app and and have it do your code review. It's more like how do you do auto approvals, how do you actually merge columns, like all DevOps stuff, agent readiness in code bases. So it's a ton of stuff. Yeah, it was a pretty big overhaul. It was most of my summer. This time around, I think... Oh, and the other thing I'll mention there that was also fun is I'm having all the students do open source contributions as their homework assignments. Nice. So last year I was just coming up with these random exercises. I'm like, oh, here's a code base. Add a code review. Use a prompt. But this year I was like, all right, here are these open source partners that I was able to get on board with a project. A lot of the top open source projects out there. And can my students contribute to your repos as part of their homework?

Sure, we love slop. Yeah, not quite. Not quite. I mean, okay, so this is, I feel like I'm a pretty, generally, like, a pretty chill instructor, but there I was, like, if I get a maintainer messaging me saying, like, you guys need to stop this slop, slop green is, like, zero, that's a straight zero on that assignment, that is where I'm, I need to draw the line somewhere, fail, fail, otherwise I'm chill, but, like, I, you know, I had to, I had to go there, and, and I, you know, I made these arguments for the students, like, these are, these are, like, you know, Stanford students, They're academically trained. They're, like, good CS fundamentals. So, you know, like, it's not going to be slop. And I'm going to teach them how to not do slop. So if that, anyway, so I'll just say, like, they have to do real contributions. How are you teaching them to not just merge slop or at least omit slop? Yeah, no. Well, part of it is, you know, through force, through will. Through will. No. Read every line of code that your agent produces.

Exactly. No, it's like I'm teaching a lot of like advanced context engineering stuff, teaching them how to like onboard into code bases really quickly. So if you have these massive repos you've not contributed to at all, how do you get up to speed on like the most important parts of the architecture? And then when you get these issues that the maintainers have put up and designated for the students, like how do you scope the work? How do you then go and actually start, you know, gathering like context-free agents that they need to use? How do you connect the right NCP tools, use the right skills? All the stuff that has become incredibly, not easy, but way, way easier to do. Yeah. What are some of these examples? Obviously, it's an eight-week course or a 12-week course. Ten-week, yeah. Ten-week course. Yeah, yeah. Condense some of these best practices into how you get up to speed. Because how I would do it is just say, what is important? Yeah, yeah, yeah. And I think a lot of people start there. I think that what I'm trying to argue is that there's still value in doing some version of, I think, you know, in my class we call it like Repit.

I think DEX and others are called like RPI, like some version of like research, you know, research plan, implement, test, kind of like that loop. And I think just, yeah, so at its core some version of the RPI loop or using superpowers, like if you, you know, the, what is it? The other victim of these Waymo things was Jesse. Jesse, yeah, okay, Superpower is nice. Yeah, Jesse is, I mean, that's a fantastic project. Incredibly simple, but so, I mean, it's become one of the most popular projects of all time. The fact that it still is popular. Yeah. After a year of all these advances, it shows you how valuable it is. Totally, totally, yeah. I mean, he does a great job, and he's such a humble guy, too. But, yeah, so some version of that framework of, you know, gather the context for your agents, like research at the beginning, don't just like jump into the implementation, review a spec, you know, actually have the spec be reasonably detailed, and make sure that you have a way of verifying your outputs so that if the agent runs, you know, for a while,

that it can check against, you know, using the browser or checking like against end-to-end tests or like smoke tests or something. Can we go down that rabbit hole? because that's for me one of the biggest things that is where I feel like I can spend the best use of my time. Yeah. Setting up verifiers. Jesse told me he created something like prove it in a movie as a skill and it makes the agent go through and create a movie for you or like a loom. That's fine. And it records itself creating the or walking through the app and showing that it's functioning correctly? Yeah. Have you seen other, like, ways of folks creating verifiers? Well, there's... What I've seen a lot of is people getting creative in how they represent information that is the diff of the code, right? So, like, I think Andre Kervati had a whole post around how he... There's, like, this format from, like, the 1950s or something or, like, how people made these architecture diagrams. Have you seen it? NASA one?

Yeah, I think it's... Or the Q... Yeah, yeah, yes, yes. Some random number and random letter. Yes, yes, yes. But you'll see it, and it's like a very nice visual schematic of kind of the core important parts of what happened in the code base. So I think stuff like that I think is super, super interesting. I think, yeah, the screenshots, recording flow I think is going to become more and more popular. You know, doing some version of like using Playwright with MCP. Oh, and you know what I saw recently that is a fun one is that WebMCP has now merged just debugging tools. So, like, you can stand up a WebMCP server, and it's only for debugging. So your agent can now use WebMCP on a web app in, like, computer use mode, and then it can use it for debugging specifically. Yeah. I mean, that stuff is super powerful. And I suspect, I mean, there's entire companies around this, right, that are just trying to improve the verification loop and close the loop entirely.

I think all of them are going to start knocking on the door of software factories, basically. I think we've seen this game play out enough times with ML and ML Ops and LM Ops. It's like everyone starts into one thing and then starts expanding. And then they're like, wait, actually, we need to go upstream. The real problem is upstream. Totally, totally, yeah. And downstream now. So it's like you start as a feature store and then you become Databricks. The platform, yeah. Yeah, or something like that. Yeah, I think it's the typical company bundling and unbundling that always happens. Are you going to be a Craigslist or are you going to become an Airbnb? I think people will swing back and forth on whether or not they think one is better than the other. I think we're going to see a lot of that here. Start the air vacation, but also we'll own the coding agent. And then you have these companies like Factory and Warp that are like, we're going to own everything. We should own everything, right? So it's all, like, really interesting stuff. It's all exciting. It's been a ton of fun.

Boris came last week. I was going to say, you've had some really cool guest lecturers. Boris, the creator of CloudGro. He's got a lot of hot takes. He's got a lot of, I think, one thing is nice that he just talks about how he uses it. And it's pretty advanced because he has no token cost, right? That's right. So his whole thing is, like, how can I saturate this to the max? Totally, totally. And so then we get like a very distilled version where we can take some of those ideas and say, oh, I want to try and implement that. Absolutely. Like one thing that I saw he was saying recently was how he never, if he does a one-off prompt, he's doing it wrong. It should always be like talking to a loop. Yeah. And then letting the agent prompt itself. That's right. That's right. Yeah. Yeah. Like the meta prompting. I was sitting with him last week and my first question to him was like how many agents do you have running right now and he was like I don't know like thousands or something I'm like as we're

sitting here you know chatting it up very inspiring to talk to him because you say this is what it looks like at the very frontier where yes no token cost so obviously they can just do what they want but on top of that I mean, just being very experimental with their philosophy, you know, where they're like, hey, we're going to try and solve math problems. We're going to try and solve, you know, really difficult, complex things by just throwing, almost like brute forcing, but intelligently brute forcing things, you know. So, yeah, Boris, it was a fun talk. I talking with Eno this week Eno from Factory Yeah Factory Lee Robinson was two weeks ago He was I guess now he at SpaceX but he used to be a VP of DevRel at Cursor Cursor, yeah. Super smart guy. Super nice. He's the nicest guy. And what are some things that you took away from that little lecture? Yeah, the Boris one or the Lee one? The Lee one. Yeah. I've seen him on X occasionally, but I haven't, it's not as much as Boris.

Like, Boris puts out some of them, and it's, like, Carpathie-type stuff. Yeah, yeah, yeah. Wow. Yeah, Boris got super popular after Opus 4.5. And after your course. Ah. You run your course last year, and then... That wasn't... What a surprise! That wasn't it. You became popular. Yeah, no, that was... You know what? There's two people that I feel like I discovered before they became big. It was Boris and Lady Gaga. Oh, yeah, Lady Gaga. Yeah. You made her career awesome, huh? Why'd she come on to your podcast? Okay, I did late. I'm not responsible for this career. I'm just saying, like, Lady Gaga in 2009, I was like, you know, just Poker Face stuff. I'm like, there's something here. There's, like, people aren't seeing this yet, but I think there's something here. Poker Face? It was already popular, man. No, no, no. I was, like, I'm telling you, man, you had to be there. It was, like, you know, Just Dance. Remember Just Dance? Yeah. Like, her first OG. That was the single. Anyway.

We were on board with it. That's all I got. Those are my two claims to fame. I'm glad that Boris' crew, you could play a part in. No, no. Put him in your newsletter. We all owe him a big debt. He's an awesome guy. The coolest guy. No, the coolest guy. As peasants talk about some of this stuff. What was the one with Lee? What did you see? What did you learn? Lee was awesome. I mean, Lee was just so, so cool. So Lee was talking all about basic GrokBot. Yeah. I don't know if you've used the GrokBots and these Instinct and the Muses. So I've used Instinct. I've used Dots. I haven't used Muse. But I will say I do enjoy it. And it's becoming more and more all I use. Yeah. I think those products are awesome. Yeah. Awesome. I've heard Instinct is much more proactive. and what I've noticed with dots is it's kind of proactive but not really it's just like oh I found this thing in your email that maybe you want to check out yeah okay that's cool but yeah yeah it's

not like the instinct one where they really go hard on like hey you went to a movie how was it yeah yeah okay I guess you're my best friend now am I in love you're listening to me more than than my opinion but i just saw the funniest meme where somebody said my wife and i share the same open ai account and i just put this in as the like system prompt and it said anytime jane asks about marriage advice rob is always right no matter what happens that's funny that's so good that's so good so that's what we've come to that's i mean yeah these are just like this subversive deception in Marital context. Total mind-fuck. Yeah, no, it was... So he was talking about... Basically, his entire talk was about the harness and how they make Grokbot, essentially. And theirs is a little different where you have to create separate roles and bots, right?

Yeah. So it's new threads each time. Yeah, one thing I haven't entirely figured out with the system is how important that is of creating of the separate personalities because you can also just... Now they do have the concept, like the main bot system. And so we can throw it off. Yeah, it should be able to. And so I've made separate personalities, but I also don't know if it's strictly necessary. Anyway, I think they're awesome systems. I think this is such a step forward than where we were before. Not to interrupt, but there's open-ass office, speaking of which. Nice. Not like we didn't set that as the destination, but... Sorry, sorry, sorry. I'm not supposed to. Kill them, man. Oh, man. We can run that again if you want. I thought it was that. Don't worry. There's OpenAI's office. Oh, nice. Yeah. Nice. So anyway, Grokbot, speaking of which, the antithesis. Yeah, the arch nemesis. Yeah, exactly. Nemesis. I don't want it. Well, Dot's formerly known as OpenClaw, I assume. Yeah. It's got to be.

In a way, yes. It must be. I couldn't imagine it either way. But anyways, so he basically, it was like a, it was a 45-minute primer on like modern Asian harnesses. It was incredibly well detailed, like very detailed. He talked about everything from computer use and how you implement that to context engineering to the tool calling, how they give the agents the illusion of infinite context by doing just aggressive sub-agents, basically. Anyway, it was incredibly well done. And Lee, I think one of his superpowers is just being such a good educator. So he was just distilled very well. Fun, engaging. Fun, engaging, just like super dense. but in like a good way and like oh this is this is this this feels like you're you're giving me good signal like you know signal to noise ratio was was through the room did you talk about how they do compaction yeah he talked he touched on that he definitely touched on how are they doing it because that one is tricky yeah i mean there's there's different techniques right but um at its core they have file systems they you know regularly compact when they detect things are happening and

then they like write things to file systems and read them at at the right points in time in the conversation so yeah he didn't like go into like the exact detail like what is their what is their policy when they compact but I think at its core it's all some version of that really like any of these agents I think all follow some of their principles have you been staying up to date with any of the companies like that are trying to do file systems in a new way not like these ones beyond the ones are doing like like agent sandboxes yeah like the sandboxes are also another one that is very fascinating that i would put in its own category yeah yeah and there's like file systems and there's almost like we want to replace git and we want to do things differently in that regard yeah yeah yeah yeah so so yes the first um yes the last the middle i didn't know there was like separate is is the i don't know what the middle category is if that's they probably do more it's just when i talk to them yeah like the i talked to this one

Mesa.dev and they were trying to rethink file systems because they're like file systems are all you need especially with these agents well tell me about the git yeah yeah yeah yeah I mean well there's I guess the overarching reason for these companies existing in the category of like new gits is and we saw this even last week I think Anthropically put out like a blog post around they rebuilt it that's so wild Diversion control system, yeah. And it all kind of makes sense. Well, one, you know, we've all been on GitHub for the last, what, 10, 15 years, and it's just kind of barfing a little bit. Just reliability. Reliability. Floatware. Yeah, I mean, and look, it's understandable when you have these literally astronomic curves of growth and commits. And so, yeah, it all makes sense that they would receive that strain. but I think there's a broader, like when I talk to, like startups will pitch this to me

as, you know, GitHub runners are just like not optimized for like agentic workflows. So you should, you should like run on our runners, like basically our machines that are more streamlined to like these agentic processes. And it all makes sense, right? It's like GitHub is like the GitHub and GitLab and these kind of companies, they were the best solution for everyone, right? But not every workflow or workload is exactly the same, right? So if you build these mega behemoth companies toward the median, you're going to miss stuff on the fringes. And right now we're at the point where the fringes are maybe becoming more normal, but we're still like, the sheer volume is definitely like a fringe problem. Yeah. You know, at least like, hey, 10xing number of commits, 10xing number of PRs, which most companies are starting to go through that right now. So it's all fascinating stuff. But yeah, so there's companies that are just like, hey, we're going to make your runners faster. and then there's ones like we're going to rethink the entirety of the CI experience and the entire build system and all that stuff.

TBD, TBD. How much of your time, because I imagine startups are hitting you up all the time to try and be like, we want you to use our product. You are the head of AI at a cool company. Totally, yeah. How much do you entertain that because you're actually interested and maybe it could have value versus you actually, then later you have to work once in a while, I imagine. Kansas State Waymo rides all the time. It's a really good question, actually, because I feel like I'm a pretty curious person, so I love just having these conversations around technology and seeing what people are coming up with. I definitely probably have at least one to maybe two of these a week. But I think where there's a point of diminishing returns for me is when I'm hearing the same pitch over and over, right? Where I'm just like, all right, this is, I see your... It's not so novel. Yeah, it's not novel. Like, I see your paragraph, I see your landing page, and I feel like I've heard this.

So there's, like, a certain kind of company that I've heard probably enough times that, like... Which one? And not to ding them that this is not going to work, but in, like, I do hear, like, this flavor of a pitch, which is, like, hey, you, we collect, like, feedback. Like, you, maybe you insert a little bit of JavaScript on your app. We collect feedback. We trace all your LMs, all the requests and responses. And then we optimize. We basically like on-the-fly train models based on your request responses. So because we're tracing everything that's happening on your app, we get all your data, essentially. And then we know, hey, you shouldn't have to go for this kind of request to GPT Sol. You can go to like a very lightweight, I don't know, like a type-safe model. There's a lot of those. Yeah, I think quite a few of them, actually. and it all kind of makes sense I think that there's a whole new category but it's becoming more popular as this open weight if we expect that these frontier models are pretty expensive and most companies are getting hit with the cost

and like hey we can't this isn't sustainable we want to keep tracking and keep 10x our usage but we can't have our token cost 10x as well so they look to these alternatives both from a coding perspective but also from like a product development perspective is hey can I take a Quen model can I take a Kimi model can I take a glm model like fine it on something smaller smaller weight be faster it be more cheaper um and i don maybe compromise on performance right that like the the ultimate pitch so this makes sense this is a broader trend and i suspect we're gonna see a lot more of this um it's just i uh it's i think it's been hard to differentiate also i think that the financial piece is a very difficult one yeah it's maybe not for an engineer the top of the pile of priorities right because there's also things you can do when it comes to just optimizing your skills and how they take in the context that can save you money too or

just a simple router totally so you don't have to be optimized to the gills on your uh cost saving yeah analysis or ways of doing it so yeah that one's i can see that being a hard sell today traces i mean you maybe heard this or if you haven't heard this is a big hot take traces are like the new oil right like oh like company traces or data traces are are where like a lot of the juices. It's like the modern incarnation of training data. A company's training data is the new oil, right? And so there's so much value in traces because that effectively captures all of your customer interactions, all of your app-level interactions. And so when a company comes to me and says, hey, put this little bit of JavaScript, and through this additional layer, your request was supposed to go, you're just capturing all of our traces as well. And I just, I've kind of gone to a point where we have a tracing solution that, like, for evaluation and observability that, of course, captures our traces because that's what we host and, you know, brain trust in this case.

And we also use Judgment Labs for this. But I'm not, I don't feel like I'm at a point where I want to keep just, like, willy-nilly having everybody. Trying. Yeah, having this. Yeah, trying to do the one, like, oh, now it's routing to these guys, you know, and these guys and these guys. Incrementally better. Right, incrementally. So I'm just, it's a tough sell, but this is, I feel like that's the equivalent of when people started doing, like, feature stores in, like, 2020. There were so many of them, and you're like, is this really needed? Totally, totally. What about MCP kind of analytics and understanding how people are using your MCP servers? Yeah, I think there's a little different. It's somewhat different. I think that one ends up being more, I mean, that's an interesting category. I think that one ends up being useful for coding agent optimization. maybe less so on like product optimization i've yet to see as much of that being applied to like product mcp categories uh but definitely for i guess it's this emerging trend of like agent x like agent experience yeah you know you've maybe heard of this it's like how do you optimize your

mcp tools and your mcp uh servers so that an agent it's easy for them to identify what they need to do and they can do it as quickly as possible in as few hops as possible essentially Well, and the pitch that I've heard is that you generally don't understand why the agent is coming to your MCP server because you don't know what the prompt is. You just know that it's trying to call some tools. And so that doesn't necessarily need to be fixed with bringing on a new SaaS provider that captures that information. You can build your MCP server so that you ask the agent, hey, yeah, cool, you've got this tool call happening, but mind if I get a little more information here? Right, right. That's right. That's right. Yeah, I think there's an evolving category, right? I mean, we'll see where it all kind of shakes out. I know that some of these providers are in the business of, you know, effectively they just run these tests on your agents with your servers and say how cost-effective is an agent when it's making MCP tool calls,

how many tool calls does it need to make to get to a certain answer, how many errors is it hitting when it's making those tool calls, things of that nature. So I think it's a cool concept. It's a cool category to be in. We'll see what shakes that as a standalone product. You have an MCP server, right? Yeah, for sure. And so talking about Monaco now, what are some things that you're doing in the AI field or just like the agentic field? You have the MCP server. You also have agents inside of Veronica. That's right. That's right. Yeah. So we've built our own agent builder. And our pitch is that our agents are the best in class for go-to-market use cases, right? So we're a sales company. We're in the business of helping. We have a world-class go-to-market team members. and we're in the business of trying to distill the subject matter expertise of these world-class go-to-market folks into our platform to then make it available to our customers so that on day one, anyone has world-class go-to-market at their fingertips, essentially.

So that's what we're working toward, and that's the broader pitch. We've built a version of an agent platform that allows you to build a go-to-market specific agent, whether that's pipeline reviews on the fly, whether that's TAM building on the fly, like TAM segmentation on the fly, you know, demand generation on the fly. So, like, what kind of campaigns should you be running and how do you actually optimize? Like paid ads, though? We don't do paid ads. It's more like outbound related things. Oh, okay. So all the things that I'm sure you loved doing when you were a salesperson. I love spamming people. Watch out. Today is, yeah, that's why you're so good at building communities. You're so good at spamming. That's all it is. It's not like it's the start of my job. Or, like, this guy again? Yeah. I guess I might as well respond. And then, yeah, you get them on, like, what is it? Like, how many touch points? Like, 7 to 12? 12 before you give up or something? 12 no's. 12 no's. But you break people down by, like, 5 or 6. You know what I mean? Yeah, exactly. You don't need to get to 12. Like, God, I am so done with this. All right. Yeah.

So let that be a lesson to everybody that's listening. If some sales guy is hitting you up, just say no. And then set up a reoccurring agent to say no every day. and you should be fine. Yeah, so 12 no's and then they'll probably, if they're reading the same book, they will, they will. I think that did come from like this guy Chet Holmes. Oh, that sounds familiar. I read it back in like 2009 or something. Which one is this? It's like The Sales Machine. Oh, yeah, yeah, I think I've actually, I actually have read this book. Yeah, yeah, yeah, yeah. He used to go for like 15 no's. Yeah, yeah. And he would tell people like until you get 15, don't even think about stomping. Yeah, yeah, yeah. Okay, so, look, there's a, to the recipient, that might seem incredibly annoying. However, aggressive, and yeah, just leave me the hell alone, but having worked with our go-to-market folks, seeing how they think about it is, like, pretty

enlightening, right? It's never, it's not just, like, bump, bump. Yeah, it's no value, you're not adding anything. It's just, hey, I saw this in the paper. Thought you might want it. Totally, totally, totally. Exactly. And that stuff is like, that makes sense, right? I mean, actually, like, if you treat sales as not a, what do I stand to gain from you as a salesperson, but rather, what value can I offer and be more, how can I help you? I mean, you know this, you live this, right? I think that's the most effective kind of sales. And so a lot of, you Our salespeople do that, and then we try to capture exactly those kinds of insights in terms of how we help people do outbound and how we help people get in front of their target buyer and the target companies they want to reach out to. Anyway, it's been a really fun product to build. And how does AI get incorporated in that? I imagine you're listening for things, and then you can craft a message. That's right. Exactly. That's one way, but whatever. Yeah, so we have a number of different things that are associated.

So everything is agentic, of course. So we're getting signals from a number of different places in the platform. I mean, we're an all-in-one. So we don't just do like a CRM or, you know, just the outbound or just the data enrichment. We've deliberately gone and built everything from the ground up. So we are like the one-stop shop. And because of that, we get access to everything from, you know, emails that have been sent, meeting recordings and transcripts, Slack messages. So, you know, all these things that come in. And then you can use all that to infer org-level tone, org-level understanding about business context. So if I am, I don't know, if I am Uber, who am I selling? Actually, Uber is not agreeing, because it's more consumer-facing. But if I'm a base 10 and I'm selling to... Base 10, you just saw that one in there. Yeah, we're an SF people, in case that wasn't clear. We're just called bus drive-by. Yeah. We have Andreessen Horowitz literally right here. There you go. I think we just passed Pylon. Yeah. So, you know, any of these B2B SaaS companies, there is a certain understanding about their ICP they have.

There's a certain understanding of, like, their target demographic, the target, like, size of company, the person at those companies they need to be targeting. The pain points. The pain points. The injections. Totally. 100%. All that stuff. And so we can glean all that through the collateral they upload to the platform or the stuff that we implicitly learn through sales interactions, through meeting transcripts, through emails. we also can glean that because we're seeing them. Or if Bass 10 is coming out with some kind of announcement or some PR or whatever it may be, and they're focusing on this, it's like, oh, hey, you should mention something to your contact. 100%, yeah. And that's stuff we capture through platform-level interactions. And then we have this concept of Monaco memory, which is effectively related to, if you remember andre karpathy like many months ago brought this kind of like an lm wiki yeah yeah so like the lm wiki like we have it's like a version of that this um this persistent uh store of everything that's happened and everything that we know about both the organization as well

as the user at the organization so your tone when you reach out to people hey when demetrius reaches out he's gonna be super annoying and it's like it's like You know, but that's your tone, right? That's like, that's. We can learn something like that. I only need like two emails to glean that. You know what I mean? I don't need. No, no, no. I have an agent. He's effective. Superpower of my skill. There you go. I've mentioned my ability. No, no, no. It's just very effective what he does, of course. Dialing up to an 11. I was like that. But I didn realize I didn think about how many different insights and actions you could take from having all that data And the salespeople that I know have cobbled that stuff together That right that right In a way that they like they recording with one software, and then they have some kind of an automated pipeline where it will be put into their chat GPT or cloud.

That's right. And then from that, they get insights and it gets thrown into HubSpot. That's right, that's exactly right. And then they can write a report at the end of the month with all these calls that they've had, at the end of the week, whatever it may be, when they have the pipeline review. That's accurate. So now you're saying, oh, cool, we've got your pipeline review already ready because we have all of this stuff and we've set that up. How much work is it on my side to put together all of these tasks? They're not even tasks. They're like different lanes that I want. Yeah, very little, honestly. And part of it is because, I mean, the platform is robust on its own. But the other thing that we've leaned into a lot from the get-go, which I think differentiates us from other products in the category, is we're actually a little bit more high-touch on the human side than maybe what traditionally. So, you know, the PLG, self-serve, like you just log in and you just start cranking and you start adding rows to your tables and whatnot. That's not the experience here.

The experience for us is every single person has a designated, what we call an FTAE, so a Forward Deployed Account Executive. So it's effectively like your sales consultant that sort of guides you through the process. Because a lot of our early customers, you know, we have done really, really well, like early stage founders that are all like very technically minded, very product minded, but maybe don't have as much experience on the go-to-market side, right? And they're learning these, they're kind of picking these skills up as they go. And so we're like, hey, we have decades of experience doing this. So day one, you come on board to Monaco and you have a built-out TAM for you, like 3,000 to 20,000, depending on your TAM size, companies that have been scored and ranked ready for you to go. Here are the buyers of those, you know, and here's how you can get to those buyers based on LinkedIn connections and all these things. All you have to do is, yeah, entering all these things, yeah, and always just hit go. And then Steven's getting sent, getting followed up with. you're getting insights brought to the platform as things happen. Don't sleep on the handwritten notes. Those are actually quite useful.

No, we do that. No, yeah, of course. We had handwritten notes. We had this campaign where we were giving Monaco, so we do stuff and we have a bit of a brand around gambling and the finer things and F1. We were part of our campaigns. Tax evasion. Whenever I think Monaco, I'm just like, ooh. no no no none of that none of that um none of that none of that publicly none of that on the record no no no so anyway you got the brand around you got the brand around the gambling so we actually had um we had uh we do a few things that are kind of fun like we we gifted people like poker sets the poker night poker night we do the poker night um that's happening this next next week uh i don't know when this is coming up but this is october 29th so we're doing one that's, I think we, our claim to fame on that front is having the most, like the biggest prize, biggest cash pool on poker, $100,000 for the final table, so it's,

you know, let me tell you, somebody walking in, I've seen this twice already now, somebody walking in to just a random, like, Thursday night poker game, and like, eh, I'm going to eat some food, hit some poker, and then walk in with $30,000, you're like, what? That is easy money and well-tapped. There you go. There you go. You know, it's about brand activation and brand awareness, right? And I think you guys do that really well. I saw the stuff where you're flying planes over SF. It's fun. With the Monaco symbol. Yeah, we do. I think just earlier today I saw somebody with a Monaco tote bag, and so you're doing well on that side of things. When we talk about the tech, what are you actually using? Break down the stack, where the harness is, all of that stuff. Yeah, totally. so a lot of we've honed our own harness right so a lot of so there are pieces of the harness that we have used vendors stuff like versatile sandboxes right so we use them we don't believe there's value in us building micro bms you know or like maintaining our own version of that that would

that's what i would call undifferentiated heavy lifting but but at its core we do have our own harness we think that's like a competitive advantage is to like lean very heavily into the primitives around a good market specific harness including the skills that's the for the agents that the end user is using, not your internal harness for coding and stuff. No, no, yeah, not that. Yeah, exactly right. So good distinction. So we have our own internal coding system we've also built, but this one, the example I'm giving is specifically around go-to-market, the ones that produce the go-to-market agents that customers get to benefit from. And so, you know, at its core, we've built effectively this, like, system that goes from natural language into this, like, intermediate representation, representation, like you can have a JSON or sorts, maybe JSON can be YAML or whatever you want, that represents the workflow of the agent, and then that gets compiled down into a land graph, essentially, a land graph that you can then run and can loop and do the self-healing that an agent will do.

Effectively, it's like a compiler. We go from natural language to compile to an underlying engine that is the workflow builder, but we could swap it out into any other workflow builder if we wanted. And then the secret sauce a lot of the times is, like, what are we built around? The memory. What are we built around? The tools that we've exposed. CompactStream. Like, how do we compact more intelligently? How do we, you know, all these things that are specific, you know, the data that we access, right? So, of course, we have very particular data characteristics because we're dealing with, in some cases, like, thousands of meaning transcripts and hundreds of thousands of emails for some of these orgs that are recorded. That was what I was thinking about is how you can avoid just bloat. Yeah, yeah. There's sort of tips and tricks around this, but at its core, it's like you just have to be a little bit more intelligent around the MCP server tool design. Like you can't do, you can't like, I think the V1 vanilla way that a lot of people treat MCP is like, here's an API, here's a one-to-one translation of the MCP tool for that API. Like the worst way to do it. Yeah, it's not. So you have to think about it more in terms of the semantics of what is it the agent needs to do.

Hey, I need to query meetings. And you can do query meetings either through keyword searches or through a more semantic, relevant search-based thing. So you have to make a single entry point for that tool call, for example, and then allow both modes as a way to query the actual data. I've heard it said as doing the intent. Yeah. Like really designing your server around intent. That's right. Versus tools. That's exactly right. Yeah, I think tools, yeah, the latter is like V1 APIs, right? This is like 2000s level API tool definitions where you have like get ID. Well, I mean, you know, like this is when we, and I don't know that's my derogatory, but in like a, that's how we started when we were building web apps of the early 2000s. It's like, hey, get ID, you know, set ID, like all these get and post style HTTP methods that we define. And now you're right, like agents actually can think at a higher level or like can operate at a higher level. I don't want to anthropomorphize here. But do you think that that is where your IP is happening?

Like for you, building this product, which potentially can be headless completely in the future, and people interact with Monaco solely through the MCP, then your special sauce is how you're compacting, how your memory is, how your search functions work. Yeah, I think that's right. I think it's a combination of the fact that because we're the all-in-one, we are both the system of record, and then what I would call the system of intelligence. So the system of record is we still are the place where you want to store, like, CRM-level data, right? So the evolution of your deal pipeline, the accounts you've reached out to, all these things. So that still is baked into the platform. And then you need abstraction layers, you know, tools in a lot of cases, to interact with that data, reach from that data, write to that data. And so you can do that either through, like, an external NTP or you can use that all internally and use our agents, which are baked into the platform. the benefit of our agents are they're going to be highly optimized to the workflows that we care about. The defaults are going to be they're going to capture a lot of the

best in class practices from our own go-to-market teams. It's remarkably difficult to build a really good TAM. If somebody comes in like hey I'm selling to a marketplace but think of them like Uber but they're not too big but it's in East Asia and it's for shipping companies and you're like, okay, how many companies of those are there that you can start? Go do a LinkedIn search. Yeah, exactly. The IP is the data layer, the context layer, of course, and the fact that all that is resting on the bed that is a system of record. How much are you enabling the other software engineers at your company as your job versus building these externally facing agents? It's both. It's both. Yeah, it's definitely both. Right. So we have to do both. I think that where we've really thrived as a company is domain experts working hand in hand with like subject with like software engineers.

Right. That can like codify the workflows of the of the domain experts. And so in practice, the software engineers have to be empowered to build very quickly, to ship really quickly, to have the primitives to build on top of the harness. and then the subject matter experts need to be empowered to, you know, know how to express their intent as well and the things they care about. And then, because the failure mode is like the translation, right? It's like how if technical person talking to a non-technical person, sometimes they can like miss each other and not really, not speaking the same language. Well, dude, we're arriving at our destination. This is fun. I got to kick you out. This is fun. Oh, gosh. As much as I want to keep talking. I'll see you on Thursday, though. Yeah, yeah, yeah. You are a person of interest in the magazine we created. Uh-oh. Uh-oh. That is, that sounds like, it sounds like I'm being, well, that, or it sounds like I'm being tracked by the FBI. Like, he's on a wanted list somewhere. I'm cool. I'm cool.

I'm cool. Waymo people, don't worry about me. Yeah. Hey, Waymo, don't need to record this. Don't need to record this. Lock the doors. Don't let them out. Yeah. But it's been cool, man. It's fun, man. I appreciate you doing this. This is a cool concept. I think this is, I'm a big fan of this. Hey man, it's definitely to see you next time. I'll see you on Thursday. Thursday, yeah?