Sam Whitmore: So this is another experiment that we ran last year and didn't decide to ship at that time but may come back to LM judge but one that was also agentic and could write code. So it wasn't just picking but also like taking the learnings from two models or and models that it was looking at and writing a new diff. And what we found was that there were strengths to using models from different model providers as the base level of this process. basically you could get almost like a synergistic output that was better than having like a very unified like bottom model tier.

Jonas Nelle: We think that over the coming months the big unlock is not going to be one person with a model getting more done like the water flowing faster and we'll be making the pipe much wider and so paralyzing more whether that's swarms of agents or parallel agents both of those are things that contribute to getting much more done in the same amount of time.

Interviewer: This week, one of the biggest launches that cursor has ever done um is cloud agents. I mean, I think you you had cloud agents before, but like this was like well, you give cursor a computer, right?

Interviewer: Um so is this basically they bought autotab and then they repackaged it. Is that what's going on or

Sam Whitmore: that's a big part of it. Yeah. Cloud agents already ran in their own computers but they were sort of site readading code. And those computers were not they were like blank VMs typically that were not set up for the DevX for whatever repo the agent's working on. One of the things that we talk about is if you put yourself in the model shoes and you were seeing tokens stream by and all you could do was site readad code and spit out tokens and hope that you had done the right thing.

Interviewer: No chance.

Sam Whitmore: I'd be so bad. Like you obviously you need to run the code. And so that I think also is probably not that contrarian of a take, but no one has done that yet. And so giving the model the tools to onboard itself and then use full computer use end to end pixels in coordinates out um and have sort of the cloud computer with different apps in it is the big unlock that we've seen internally in terms of use usage of this going from oh we use it for little copy changes to no no we're really like driving new features uh with this kind of new type of aentic workflow.

Interviewer: All right let's see it.

Sam Whitmore: Cool. So this is what it looks like in cursor.com/agents. So this is one I kicked off a while ago. So on the left hand side is the chat very classic sort of agentic thing. The big new thing here is that the agent will test its changes. So you can see here it worked for half an hour. That is because it not only took time to write the tokens of code, it also took time to test them end to end. So start at dev servers, iterate when needed. Uh, and so that's one part of it is like model works for longer and doesn't come back with a I tried some things PR but a I tested it PR that's sort of ready for your review. Uh, one of the other intuition pumps we use there is if a human gave you PR, asked you to review it and you hadn't they hadn't tested it. You'd also be kind of annoyed because you'd be like only ask me for a review once it's actually ready.

Interviewer: So that's what we've done with uh simple question I wanted to gather up front. Some PRs are way smaller like just copy change. Does it always do the video or is it sometimes?

Sam Whitmore: Uh sometimes.

Interviewer: Okay. So, so what's the judgment?

Sam Whitmore: The model does it. So, we uh we do some default prompting with sort of what types of changes to test. There's a slash command that people can do called slashn no test where if you do that the model will not test but the default is test.

Jonas Nelle: The default is to be calibrated. So, we tell it don't test very simple copy changes but test like more complex things. And then users can also write their agents.mmd and specify like this type of you know if you're editing this sub part of my monor repo never tested because that won't work or whatever. Okay. So pillar one is the model actually testing. Pillar two is the model coming back with a video of what it did. We have found that in this new world where agents can end to end write much more code reviewing the code is one of these new bottlenecks that that crop up. And so reviewing a video is not a substitute for reviewing code, but it is an entry point that is much much easier to start with than glancing at some giant diff. And so typically you kick one off, you it's done. You you come back and the first thing that you would do is watch this video. So this is sort of a you know video of it. Um, in this case, I wanted a tool tip over this button. And so, it went and uh showed me what that what that looks like in in this video that I think here it actually used a gallery. So, sometimes it will build storybook type galleries where you can see sort of like that component in action. Uh, and so that's pillar two is like these demo videos of what it built. Uh, and then pillar number three is I have full remote control access to this VM. So I can go here in here, I can hover things, I can type, I have full full control. Uh and same thing for the for the terminal. Um so you know I have full access and so that is also really useful because sometimes the video is like all you need to see and often times by the way the video is not perfect. The video will show you is this worth either merging immediately or often times is this worth iterating with to get it to that final stage where I am ready to merge it. And so I can go through some other examples where the first video wasn't perfect, but it gave me confidence that we were on the right track. And two or three follow-ups later, it was good to go. And then I also have full access here where some things you just want to play around with. You want to get a feel for what is this like? And there's no substitute to a live preview. And the VNC kind of VM remote access gives you that.

Interviewer: Amazing. Well, sorry, what is VNC?

Sam Whitmore: Uh it just the the remote desktop. Uh remote desktop. Yeah.

Interviewer: Uh Sam, any other details that you always want to call out?

Jonas Nelle: Yeah, I mean for me the videos have been super helpful. I would say especially in cases where a common problem for me with agents and cloud agents beforehand was almost like underspecification in my requests where our plan mode and kind of going really back forth and getting detailed implementation spec is a way to reduce the risk of underspecification. But then similar to how human communication breaks down over time, I feel like you have this risk where it's like, okay, when I pull down go to the trouble of pulling down and like running this branch locally, I'm going to see that like I said this should be be a toggle and you have a check box and like why didn't you get that detail right? And having the video up front just has that makes that alignment like you're talking about a shared artifact with the agent very very clear which has been just super helpful for me.

Sam Whitmore: I can quickly run through some other examples. So this is a very front-end heavy one.

Interviewer: I was going to say is this only for front end?

Sam Whitmore: Exactly. One question you might have is is this only for front end? So this is another example where the thing I wanted it to implement was a better error message for saving secrets. So the cloud agents support adding secrets. That's part of what it needs to access certain systems.

Interviewer: cloud is working on cloud agents.

Sam Whitmore: Yes. So this is a fun thing is

Interviewer: it can get super meta.

Sam Whitmore: It can get super meta. It can start its own cloud agents. It can talk to its own cloud agents. Sometimes it's hard to wrap your mind around that. We have disabled its cloud agents starting more cloud agents. So we currently disallow that.

Interviewer: Someday you might.

Sam Whitmore: Someday we might. Someday we might. Um so this actually was mostly a backend change in terms of the error handling here where if the secret is far too large, it would um Oh, this is actually really cool. That's the dev tools. That's the dev tools. So if the secret is far too large, we don't allow secrets above a certain size. We have a size limit on them. And the error message there was really bad. It was just some generic failed to save message. So I was like, "Hey, we wanted an error message." So first cool thing it did here, zero prompting on how to test this. Instead of typing out the like a character 5,000 times to hit the limit, it opens DevTools, writes JS or to paste into the input 5,000 characters of the letter A and then hits save, closes the DevTools, hits save, and gets this new um gets the new error message. So that uh looks like the video actually cut off, but here you can see the here you can see the screenshot of the uh of the error message. So that is like front end, back end kind of end to end feature to to get that.

Interviewer: um and you just you just need a full VM, full computer, run everything, right? Okay. Yeah.

Sam Whitmore: Yeah. So we've had versions of this. This is one of the autotab lessons where uh we started that in 2022. Uh no, 2023. And at the time it was like browser use, you know, DOM, like all these different things. And I think uh we ended up very sort of AGI pilled in the sense that just give the model pixels, give it a box. Sort of a brain in a box is what you want. And you want to remove uh limitations around context and capabilities such that the bottleneck should be the intelligence. And given how smart models are today, that's a very far out bottleneck. And so giving it its full VM and having it be onboarded with DevX set up like a human would has just been for us internally a really big big step change in capability.

Interviewer: Yeah. I would say you know let's call it a year ago the the models weren't even good enough to do any of this stuff right. So

Sam Whitmore: even six months ago. Yeah.

Interviewer: So like yeah what what people have told me is like around about solder 45 is when this started being good enough to just automate fully by pixel.

Sam Whitmore: Yeah. I think it's always a question of when is good enough. I think we found in particular with Opus 4546 and Codeex 53 that those were additional step changes and sort of the autonomy grade capabilities of the model to just go off and figure out the details and come back when it's done.

Interviewer: Yeah. Anything to add? One other shift that I've noticed um as our crowd agents have really taken off internally has been a shift from primarily individually driven development to almost this like collaborative nature of development.

Jonas Nelle: For us, Slack is actually almost like a development um an IDE basically.

Interviewer: That's why I'm like maybe maybe don't even build a custom UI like maybe like you know that's like a debugging thing but actually it's Slack.

Jonas Nelle: I feel like yeah there's still so much to left to explore there but basically for us like Slack is where a lot of development happens like we will have these issue channels um or just like this product discussion channels where people are always at cursoring and that kicks off a cloud agent and um for us at least we have team follow-ups enabled so if Jonas kicks off a at cursor in a thread I can follow up with it and kind of like add more context and so it turns into almost like a discussion service where people can kind of like collaborate on UI. Oftentimes I will kick off an investigation and then like um sometimes I even ask it to like get blame and then tag people who should be brought in because it can tag people in Slack and then other people will

Interviewer: can tag other people who are not involved in conversation

Jonas Nelle: can just do at Jonas.

Interviewer: Yeah. That's cool. You should you guys should make a bigger deal out of that.

Jonas Nelle: I know. um it's a lot to I feel like there's a lot more to do with our Slack surface area um to show people externally. But yeah, basically like it can bring other people in and then other people can also contribute to that thread and you can end up with a PR again with the artifacts visible and then people can be like okay cool we can merge this. So for us it's like the ID is almost like moving into Slack in some ways as well.

Interviewer: I have the same experience um with but it's not developers it's me designer salespeople.

Interviewer: Yeah. So me on like technical marketing vision, designer on design and then sales people on like well here's the legal thirds of what we agreed on and then they all just collaborate and like correct uh the agents.

Sam Whitmore: I think that we found these threads is the work that is left that the humans are discussing in these threads is the nugget of what is actually interesting and and relevant. It's not like you know the boring details of where does this if statement go. It's like do we want to ship this? Is this the right UX? Is this the right form factor? Uh, you know, how do we make this more obvious to the user? It's like those really interesting kind of higher order questions that are so easy to collaborate with and leave the implementation to the to the cloud agent.

Interviewer: Totally. And no more discussion of like am I going to do this? Are you going to do this? Curs is doing it. You just have to decide you like it. [laughter]

Interviewer: One more question on on just the cursor in general and then uh you know open-ended for you guys to plug whatever you want to plug. How is cursor hiring these days? Predictions for the end of the year, if you have any.

Sam Whitmore: Predictions are not going to go well. Well, one other plug. Yeah. I predict that by the end of the year the volume on I think it will take longer than people think and longer than we think for cloud and agents working in their own boxes to surpass local agents. But I think that crossover will happen before the end of the year and probably by the end of the year agents running in the cloud will be a multi like more than 2x the volume of local agents.

Interviewer: Okay, you're leaving me an opening. What's not good today?

Sam Whitmore: Yeah, there's a bunch of hard things. So, one of them is just getting those sandboxes to be really, really good. And a thing that was part of this launch that we spent an inordinate amount of time on is cursor.com/onboard where you pick a repo, add secrets, give it access to things, and the agent just goes off and installs things. Yeah, we worked a lot on that. but, um there's still a lot to do there, right? like setup one two two things maybe it's too bad it's too slow um working on it uh setup is not like a unitary thing where everything is set up or not right like things will break over time you have new dependencies you need access to new systems like you change where your database lives so that's one part of it

Jonas Nelle: And then the other part of it that starts to get really interesting is when the model starts editing its own system prompt. self-awareness to us in this context is kind of like um not like the model itself having a notion of consciousness but more like knowing like what system it's operating in and the the constraints of that system and potentially being able to have agency in like optimizing itself to operate best in the in that system.

Interviewer: Yeah, this is more abstract than I ever thought we'll get at this [laughter] tier discussion.

Sam Whitmore: No, I mean you guys are like my sort of needing example what an agent lab looks like and like can be successful and like I think people always hungry for insights into how you guys operate. So, thank you for taking the time to share.

Jonas Nelle: Yeah, thanks for coming.