Transcription
Hey, hey, hey. We are looking at a 5,000 share order. Hello everyone and welcome back to Google Cloud Live. I'm Stephanie Wong and if you tune into our last few episodes, we spent a lot of time in the terminal with Gemini CLI looking at how AI helps us write code faster. But we all know that writing the code is only half the battle. The other half, and let's be honest, usually the heavier half, is getting that code to run somewhere other than your laptop. So today, we're talking about anti-gravity. It's a tool designed to literally lift the weight of infrastructure management so you can focus entirely on building. And so to walk us through it, I have two of the engineers actually building the engine. So please welcome Kevin How and Andy Zang. Hey guys.
>> Hello.
>> Welcome back to the show, Kevin.
>> Thank you. Always good to be back.
>> Yeah. And welcome to the show, Andy.
>> Thank you. Good to be here.
>> Yeah. So, you both have been working on anti-gravity since day one here. Uh, how has it been so far?
>> It's been fun. Uh, we've been doing DevTools for for some time now. Um, anti-gravity is probably one of our our proudest biggest achievement.
>> Yeah.
>> Um, and we're really excited to to see it out in the world and seeing so many people using it and all the different use cases that have been falling out of it. Um, so it's been really a magical experience for us.
>> Yeah. How about you, Andy? It's been so much fun working on anti-gravity as a engineer by trade. Um I I love building and I think it's a really big privilege to also be building a tool for building. Uh especially building anti-gravity in anti-gravity and I think especially now that it's out in the wild for several months now. It's been really amazing to get a ton of feedback from users and see how exactly can we improve the product to make it the best it can be.
>> Yeah, it must be it must be special to work on something that you're like I use this. I am an engineer by trade and I'm building something that I can see myself fundamentally changing my workflow myself. So, it'll be great to hear how you're actually using it in your personal workflows. Uh we do have a lot of people, by the way, tuning in on the live stream, everybody on YouTube, we have people on LinkedIn. Uh a lot of people are calling out your videos, Kevin, and saying, "Hey, video on anti-gravity uh is fire." And uh thanks, Kevin. Um and yeah, just a lot of people coming in from all over the world. So thanks everyone for tuning into the live stream as always uh we'll be back week over week and probably talking more about anti-gravity throughout the year. Okay so moving on just to kind of like dive into the questions a little bit before we get into the demos. So I have to ask you both are engineers you have felt the pain of both building and deploying code. So, what are the specific friction points or nightmare scenarios in the past where you were like, I really wish we had an agentic development tool that could solve this?
>> Nightmare scenarios, man. Um,
>> I don't know about nightmare, but friction.
>> I mean, we've all had our fair share of nightmare scenarios. I think, you know, trying to avoid those nightmare scenarios in the beginning is probably how we arrived at anti-gravity or just some of the dev tools that we've been working on. Um I think specifically roughly I mean if we think rewind three years ago we were we were building autocomplete and and using autocomplete products and it was a completely different world back then and so I think over time we've been more and more engrossed in this like magical world of LLM and you know AI is doing more for developers and I think we saw a really special opportunity about a you know eight months ago or so when we started working on anti-gravity where the LLMs and agents got so good that we had that light bulb moment where it was like, "Oh my gosh, we could actually build a product that is different from anything that developers had used before. Arguably, you really didn't you could look at the code less and less and and delegate more and more to the agent." And that was one of the magical experiences where we're like, "All right, we need to build a new product and that product was anti-gravity."
>> Yeah.
>> And as we saw how models got increasingly better, um the thought that came to mind is typically when you're modifying code, you're living inside the IDE. and we had the the side panel reserved for um the agent and increasingly we're wondering like if most of our time is spent on the side panel and the agent inside of it is just more and more powerful like how can we best leverage that it's like when you're imagining like you have an entire table of space and if you're only working in one corner of it are you underutilizing the entire table and so I think the thing that uh made us zoom out and think back is like hey like what would it look like if we really made this a first class experience and so uh a alongside with or within anti-gravity, we also launched the agent manager which we're going to be doing a demo of shortly after uh talking about how we could create this first class agent user experience.
>> Yeah. Yeah. We're excited to see that in action and also talk about how ways that people can use it beyond what they typically would interact with these types of tools even if you aren't a traditional software developer, right? Um, so for those who just um, you know, are kind of learning about this for the first time, I guess just to like dive into a little bit more about your engineering background, um, before you built anti-gravity, what was sort of your guilty pleasure way of building and deploying code? Was it dragging and dropping files via FTP? Was it SSH into a server and editing code with Vim? I think we've all been there.
>> So for me, uh, I love using Vim and, uh, I did not like having to manage many different services. So I was thinking like what is the best way to bundle a lot of those. I would typically deploy things to GitHub pages because it was a two-in-one. One as the source control and number two is that I can also deploy my static pages so I can see changes locally out in the web and share a link with any of my friends immediately.
>> How about you?
>> Mine I will actually change it from what I said before. Um originally I was going to go with GCS just kind of like force push to whatever bucket. But when you said SSH, it actually made me remember I I was a big like Raspberry Pi, expose my IP to the public internet and just SSH into that thing. It had one gig of RAM and you just like shove files onto that thing and
>> it would go down every every every day basically. But, you know, it worked.
>> Yeah. I mean, hey, you can tinker and you have a Raspberry Pi, like why not?
>> It was fun. Yeah.
>> Um, amazing. Um, so last week we were talking a lot about Gemini CLI uh to to write code and now we're looking at anti-gravity. So I think some people might be thinking is anti-gravity just Gemini for ops or is it actually a fundamentally different engine under the hood and how do these two worlds of like AI code assistance and infrastructure platforms actually come together?
>> Yeah. So the way we think about uh designing anti-gravity is we wanted to make it as approachable as possible. So we don't want to narrow it on specifically one use case and obviously I think anti-gravity is amazing at many use cases thanks to the models that um power it. Um but ultimately we when we think about it it's great for coding but it's also really good for research and understanding. It's great for iterating on things such as you know your travel plans for the summer. Um or it's really good as an education tool to learn about new concepts. So for me personally, I've found I've started relying on anti-gravity not only just for coding and building things like anti-gravity um but also trying to learn new concepts so it can help with my work.
>> Yeah. Yeah. Exactly. Um I I think some people might also be thinking like okay is this just a platform as a service tool? Is it a dashboard? So how would you characterize um anti-gravity if you could do it in one sentence?
>> That's gonna be a hard one. I think that if I had to characterize anti-gravity in one sentence, I would think of it as a really good tool for thinking because typically the best thinking leads to the best work. Um, and I think that um, historically I feel like a lot of tools have been more focused on the building. Um, and I think that anti-gravity has a little more emphasis thanks to agents on the thinking aspect which leads to the the great foundation for u building something great as well.
>> Yeah, I'll add in a couple more sentences because I couldn't I couldn't squeeze it all into one. Um, I think, you know, the the product has been built to enable all sorts of builders to come to the platform and I think one of the magical parts about it is we we started with the software engineering persona and thinking about that user as someone who is willing to sort of grapple with the ambiguity of an agent. Often times you sort of have to provide guidance or nudge it in the right direction. Um but starting with software engineers was a great way you know both for us as software engineers ourselves to think about the product but then also um a set of users that was willing to be early adopters and I think we're starting to see that group grow larger and larger to include folks that are technically adjacent um you know people that might not have ever opened a coding editor or they're now approaching anti-gravity or they're able to approach anti-gravity and do tasks that you know they otherwise would never have dreamed that a coding agent specifically would have been able to do. But I think the beauty of this is anti-gravity and the platform that it sits on and Gemini specifically is a very good general use multimodal model and it's able to do a go a lot more than just literally write code.
>> Yeah. And I I actually position myself I think in that bucket of users and I was just actually watching a three-hour course that a YouTuber made about anti-gravity and it's incredible the creativity that people have to build things. like it's not just hey like I need to build a website but they're taking inspiration from other sites and they're inputting that as an image file and they're creating all sorts of ways to like include their brand assets into it. So it's really cool to see that. Um, we're going to dive into an example actually of like how to build anti-gravity and be creative and think but um, how can people get started? Do you have any advice for just how to be successful on anti-gravity? What should people watch out for?
>> Yeah. Yeah. Um, so Integravity has been used in a variety of different ways. Um, we have a lot of people that are very, very passionate about our product and to that hopefully you're watching. Thank you for that. Um, but we also want to just make sure that you're following the guidelines of of how to use anti-gravity, specifically with regard to third party proxies. Um, we've become, you know, overwhelmed almost a problem of success, overwhelmed with the amount of demand for the product and we want to make sure that people are using it in the right way. And in order to give the best experience possible, just make sure you fit stay within those those guidelines, those terms of service. Uh, we want to deliver the best product for you and we believe that anti-gravity will will be that product for you.
>> Amazing. Well, let's just dive into it. I feel like we can see exactly how to build within those guidelines and see what you can actually do successfully on the platform. So, Andy, I guess I'll pass it off to you to kind of show us the agent manager and artifacts and what all this means.
>> Yeah, let's dive right into it. Um, so, uh, right now I have the editor open within anti-gravity. Um so from here you can hop into the agent manager which is the service I talked about earlier where you can uh have a first class agent experience over here. Uh and from here I'll just go ahead and uh open a workspace. So here maybe I have uh I already set up an Olympics uh medal app over here. Now from here what I'd like to do is get a better understanding of what's been going on with the uh winter Olympics recently. Like I think there have been so many cool iconic moments like uh seeing Alyssa Leu's crazy performance and area.
>> Yeah, seriously.
>> Um and it's been awesome to see, but I feel like I haven't gotten a a deeper scoop. And so um I've loved using anti-gravity as a way to research and get a better understanding and catch up to things. So um I figured we can use the agent manager uh alongside Gemini 3 Flash to get a better sense of like the metal breakdown and maybe go from there.
>> Yeah. So, I'll go ahead and ask uh can you explain to me the medal breakdown for the Winter uh Olympics 2026? Uh so, just a super simple prompt uh asking the agent if he can explain to me what's kind of been going on. Uh I think it's top of mind and I wish I had a lot of time to uh keep up to date. Um, but luckily I think Anti-gravity is able to do a lot of the research for me, summarize things and um hopefully create a artifact explaining to me um what's been going on in the Winter Olympics.
>> Maybe it's a good time to just explain the artifact concept um and just generally what you're seeing on the screen, especially if you haven't used anti-gravity before. So this is this is our agent. This is the anti-gravity agent. And as you can see, there's what we call like tools or steps. Um, and within that, you saw Andy's first prompt, which was all right, the instruction purely, you know, five, six words or whatever, 10 words or whatever. And then all of a sudden, it's going to do all this work to accomplish said goal and including, if you scroll up, that one about searching the web. And now it's putting together what we call an artifact, which is anything that the agent puts together that helps communicate information or intent to the user. So I guess we can see that here now.
>> Yeah. So in this case, it created this artifact of an implementation plan. And before we even dive into an implementation plan, I'd love to uh I'm a bit more curious about the actual metals themselves. I don't want to interact with the website just yet. So um can you show me a table of the metal uh breakdown per country first in the plan? And so the great thing here is it's it's definitely a collaborative process. like I think of my the agent as someone that um I can talk with, I can provide feedback to. Uh and so in this case um I was hoping for some specific information and sometimes he would make assumptions about what I'd want and so um that's the beauty of of it being a conversation is I can clarify things that it might be missing just so I can iterate and get to the desired um outcome. Uh so in this case I'll just ask it if you can include in a table and as you can see here within uh just a few seconds it gone it went ahead and created a table of the winter Olympics medals. So um really cool and great for Norway to see that you know they they had the most medals this year as well as the most gold. Uh I am kind of curious what is the uh metal breakdown per gender. So I'll go ahead and ask um break this down by uh gender here. Um, and so the great thing here is that I highlighted a specific part of the table and I added a comment. And if I sent this comment um over here, the agent wouldn't exactly know what I was looking at. And so while you can share context through words, uh similar to how you're talking to a colleague or a friend, um, a lot of the things are kind of unspoken, like the actions speak louder than the words. And so the same thing here when you're interacting with the agent and its work um a lot of it is implied to the actions and not the words themselves. And in this case I was specifying like oh like in this table can you break uh this down by gender. So I go ahead and send this comment over and then uh see the agent uh iterate on this do some more research and update uh the table.
>> So what do you expect to happen here with the implementation plan? What are we going for?
>> Yeah. Yeah. So what I would expect here is I asked the agent if it can break it down by gender. So um it's going to go ahead grab the relevant data. So in this case you can see um
>> a lot of web searches,
>> a lot of web searches. It's it's working in parallel. So you know it's working really hard for you. Um and so what I would expect is it grabs all this data and it figures out like how do I blend this data into the table. And I think this is one of the powerful things about agents um that is not talked about enough is that agents are so so good at synthesizing information. Um that was literally what they were trained on where they just grabbed a lot of information, figure out how to distill it uh into this thing that can share um any kind of information looking for within seconds. So as you can see now it it went ahead and did a fantastic job of breaking down not only the gender but also the total did the math for me. Um, and so now I'm able to visualize um all the top five and maybe we'll go ahead and do a bit more iteration. So what I am curious about is also how does this compare to the past and I think that um it would be great to see you know how did how does Norway do compared to like 2022 Olympics. Um so let's see compared to 2022 here. So I'll go ahead add another comment. I think one of the inspirations for this feature was um well a like how do you interact with with people in general? They wouldn't necessarily just brain dump their their their chain of thought which is essentially what you see in conversations these days. Um and and really treating it as like all right here's a document that you're going to hand to your peer and that's what they're going to review. And in many ways, this is sort of a a GitHub or Google Docs style of commenting where you can leave a batch of comments, review it at the top, and then sort of iterate with another user uh in that way. So, we tried to model it after this and and this is one of the most popular aspects of the anti-gravity product to be able to iterate on a document as opposed to reading this thousandword, you know, often more stream of consciousness from the agent.
>> Yeah. Yeah. like as you were demoing I was thinking short of having a physical colleague sitting next to you that you're collaborating with this is like getting close to how you would explain like hey looking at this section I think we need to iterate here I would actually like to add gender uh comparisons here um so I feel like it's like we're getting as close to collaborating with someone remotely essentially and like giving it eyes and the sense of like what's going on totally so it's great yeah
>> and it's great that I'm collaborating with someone who can get results so fast uh it it this finish is so fast here. Um, and the nice thing about it being really fast is that it allows you to have so much more iteration cycles. And the more iteration cycles you have, the faster you're able to get to the vision that you have here. Um, so zooming back in, it created a 2026 versus 2022 comparison. Um, it also has a gender breakdown. And so at this point in time, uh, depending on what hat I'm wearing right now, I'm feeling like I just want to like see what the what the metal count was and then also compare to how it did historically. Um, it seems like uh Norway got more medals than a few years ago and so did the US. Um, as well as Italy. Um, let's see if there's any other important information here. See, let's see. Um, cool. This is like an interesting fun fact. Maybe there are some other fun facts. Um, but we want to do uh are there any other interesting uh highlights or tips that happened? Um, so I'll just go ahead and send this in the background as I continue reading this. And the great thing here is I can also parallelize my work. Uh, while the agent is doing some work, I can also continue reviewing. So, right below here, all of the below all of the metal counts, it also outlines a plan for how it's going to build uh this. Um, in my case over here, I already trust uh the agent. I know it's going to do a fantastic job at that. So, um, I can happily gloss over it. Um, or if I want, I could just tell it like no need to include this just to keep the plan more succinct. Um, and the great thing down here is it also shares a verification plan. And this is something that we'll talk a bit more um afterwards, which is how exactly anti-gravity thinks about testing your work and making sure that you can trust the work because it's not just about doing the work, it's also about verifying and making sure that it has proof that the work was done properly.
>> Why don't we why don't we try building this?
>> Yeah, let's go ahead and click on the proceed button. Um, so you know the nice part that we called out um, you know, you could work in parallel and I think the agent manager was designed in this way. You can have multiple conversations going at once. You can have multiple, you know, different workspaces up a workspace like kind of like a project. You can have multiple projects going at once. But we're also using Gemini Flash, which is an amazingly fast and and still amazingly powerful uh, model, especially for coding. And so this has been often on our team at least the go-to when we're building anti-gravity. Uh, a lot of folks like to use Gemini Flash because it just lets you stay in flow state a bit more than than you know just waiting for a conversation to wrap up,
>> right? But so I mean if it is a long running task you can walk away, you can grab coffee or you can do multiple tasks or you can do multiple tasks have you have no excuse anymore.
>> There's always something to do. Always something to do.
>> Yeah, exactly. Next level productivity.
>> Cool. So uh let's let's see the progress of it so far. So uh as you can see currently it's trying to implement it and it seems like it's going to start trying to test it out. So it went ahead opened the browser and the really great thing here is that it sees the fact that the website currently cannot be reached. So if we current if we go back it's going to see like okay clearly I might be doing something wrong. So how can I correct my own work? And it's going to continue iterating on this uh until it is able to find a solution or I can always jump in and try to help course correct its work.
>> Looks like it wants input.
>> Um this is another one of these things. Um obviously the browser is a immensely powerful service and we can talk a bit more about you know why we built the browser what it can do. Uh but an important aspect of this is that a browser can't access anything. And so we want to build in the right safeguards to make sure that you know if you're going to a particular URL the agent literally can click and scroll on the screen which is immensely immensely helpful but we also you know want to make sure that agent is operating within the bounds of what we expect which is why we have approvals and and things like that.
>> Um the way that I use this often is um I go half and half screen. I'll have the browser up on the left and then I'll have the agent manager on the right. Uh, and it's kind of fun to watch. Maybe we can try and resize to see
>> how the browser is doing. But you can see that there's something up there. And uh, I was hoping that we could get to uh,
>> see it actually actuate. Actuating is when it it has that blue dot and it's moving around and clicking and scrolling, but
>> it looks like it already finished up its work and and put together
>> the screen recording. So, it was a little bit too fast for us. Maybe we'll catch it on the next the next pass.
>> Yeah. Um, speaking about the screen recording, I think one thing that we can see here is that um, you can see a screenshot of this and you know how I was showing you guys that you can like select text and you can like comment on it. Um, one of the things that actually Kevin built which I think is an amazing feature is you can be uh, even more specific with the context that you provide to it. So for example, in this case it provided um, here like I can go ahead and click on this image and then I can select on uh, things that I want to comment on. So maybe let's say um I don't like this header. Can you make it super blue instead?
>> No, please.
>> Uh please. There you go. There you go. Maybe add a bit of emphasis so it actually listens to me. Um, cool. So I'll go ahead and fire this off and let it parallelize. I was going to ask if you could draw over images or sections because highlighting is great, but if it is an image, I think dragging is uh even more useful.
>> Yeah, we we I I love the image. We we also have image gen in the product which I think also inspired this. So now suddenly you've got website screenshots, you've got recordings, you've got uh nanobanana gem pickics stuff inside of um the agent. And so we wanted the ability to similar to highlighting text, wanted the ability to annotate an image. Mhm.
>> Um, and so now we can see, all right, we got that blue border. Let's see what it wants to do. It's going to start scrolling and and start screenshotting. Um, and it produces a playback. So this it's a very very neat feature. Um, and I think also more importantly, if we zoom out and think, all right, what are we doing here? We are starting to verify the results of agents much more than the actual how did we get there? Obviously, how we got there is important and that's why you can dive into the conversation if you want to. But if you're firing off, and this is our our goal and our dream, and I hope that every user watching this will give this a shot. If you're running three or four agents at once, you know, you're adding a button here, you're adding a new page there, you don't really want to or care to look at all the individual things that it did, you really just care to look at, does it work, does it look good? Um, and that's where the the verification comes in. So it'll screenshot itself and much like reviewing a a PR for example, um this is the way that we see interacting with agents and how you can sort of scale your mind space across three or four different changes that might be happening at once.
>> Yeah, that's fantastic. Um, we actually have a lot of excitement I think about the browser extension as well. Uh which I know you haven't talked about just yet, but you could actually play with this in the browser. Totally directly. So we have a question actually from Dev. Uh Ryan here uh wants to ask, "So I actually find the anti-gravity browser integration to be leagues better than some of the other similar tools out there. So what's going on under the hood to actually power this? Is it a special model training? What does the extension actually do?"
>> Sure, a lot of layers to this. Thanks for the question and thanks for the compliment. That means a lot. Um, so we worked really hard on the browser and the browser is an example of, you know, there's a lot of things that are cutting edge about anti-gravity. One of the more unique ones is probably this browser integration. I think you touch on a number of things here, Ryan. The the first of which is how is this powered? It's actually an orchestration of multiple agents. And this is one of the first times we've seen different types of models playing very well together. So for example, when you look at the conversation that there's there's sort of a main agent and in this case it's the flash model that is deciding what to do delegating tasks right and it will choose to delegate what is called like a a browser task to a what we call like a sub agent and this is basically given a goal this other agent will go off and purely try and accomplish that goal. So in this case, uh I don't know if you can see it on the screen, but it it probably said something like verify that this page or this table loaded correctly. And that's the only job of that agent. And that agent will actually use a specialized model. It's actually one that everyone can have access to publicly, but it's the computer use model. And a computer use is defined as a model that is capable of screenshotting and essentially deciding what to do next. So in this case, it probably said something like verify that the table exists. Scroll and click buttons if you need to. and it will take a picture of the screen, decide to scroll a bit, maybe zoom in on a portion of the screen and use that special capability um and then report back to that main model. So it is this orchestration of multiple things happening at once. Um this is a hugely powerful complex part of the Gemini, you know, training world as well. So it it really is a coming together of a lot of the multimodal features that you see with Gemini. Um some of it is very specific to the way that we trained specifically the computer use model. Um and it's really designed to just actuate on top of a browser.
>> Yeah. Yeah. I think it's fascinating because you know folks might not know that it is this integration between specialized model and passing back to the main model that's incorporating the feedback that it's collecting via the browser integration. Um, we talked before this this session actually about how you can use that to your advantage for testing your applications. Um, so how does that actually really affect like how people can move quickly?
>> Yeah, I have my own ways that I use the browser. I'm happy to hear you you chime in as well. I I've used it. So, on one hand, there are front-end changes where a local host will spin up. It'll click through the pages that I made. And then I actually have some instructions in my own prompts to just say, "Hey, validate that some of the other pages worked." And in this way it it you know will never it will not currently replace sort of integration tests and like the usual sort of things but it gets a finger in the wind of like all right none of the other pages are broken in any particularly glaring way. The other interesting way that I use the browser is um I have actually have it authenticated on a number of things like Google Docs for example and I will use it as sort of a a primary browser to scroll through design docs to look at um you know public documentation and because it's it's almost like the agent is looking over my shoulder because we have access to the page content I can just say hey implement this and it will be smart enough to know hey this actually means
>> the doc that I'm looking at and and in this way you bring all that richness context into the agent's execution and that that's my favorite way of using it or one of those prouser hacks,
>> right? So you don't have to copy and paste everything. You can refer to it
>> and then for like testing, you can if if you're an independent developer, it's like a team of one. You can have some level of UAT early as you're iterating so that you can just kind Exactly.
>> I think um I I'm definitely not as much of a power user as Kevin. I think a lot of how I think about it is like um what would I do? And can I just tell the agent to do that? And that's a really good limus test is like if I'm doing this repetitively, uh what would it take for me to tell the agent to just do it for me? So if I'm constantly testing this one flow, I'm like, okay, like instead of doing this thing that takes two minutes 10 times, what if I just type it out for two minutes and then save 18 minutes of my time? Um and that way like throughout every one of those iterations, I kind of decrease how much time it takes for me to just iterate on it. Um, and I think that's like my favorite way to be using the browser because at the end of the day, like I have a vision in mind of how I want it to look and um, I want to be able to iterate as much as I can uh, to it. And I rather I know exactly what are the step-to-step step-by-step flow of what I should be testing and so I can clearly outline to the agent like, hey, this is what I want you to test out. Um, and then go ahead and test it every time I ask you to make some changes.
>> Yeah. And I do know there are even more features like for any repetitive tasks that we probably won't get into today, but I did learn about skills and how you can kind of implement. But
>> you can do a whole episode on skills.
>> Exactly. We'll do that next.
>> But anywh who um that's great. Um anything else you want to show with the Olympics use case here?
>> Um, yeah, maybe some final touches here. Um I think one cool idea is like how can we visualize this not just like in a table format but if you can create a heat map of all the uh uh the countries and the medals they have. So, can you make a heat map of all of the countries um um based on the number of medals that they have? Um, so, this is going to be one last final touch that we'll do. Um, right now, it's just a it's just a usual table that you can probably find uh searching through Google, but um what I think would be really interesting is like different ways of visualizing that data. Um, I think it's a cool reminder to see like all these different countries around this world. Um, and seeing like which metals are are where. Um, so let's go ahead and ask the agent to create that kind of heat map. Um, and once again, it does a lot of the research. It analyzes the plans. Um, and then from there it figures out how exactly it wants to build it.
>> Very cool.
>> Cool. So, it's hard at work right now. It's updating all the files it made uh it made before as well as making new ones. Um, let's see.
>> So you'll notice we're spending a lot of time in the agent manager as opposed to the editor. I think that's one of the sort of high level for folks who are are more tapped into the the industry. This is a lot in many ways like the direction things are heading. these agents have gotten so good that you can sort of just trust Gemini will yeah of course verify that it it worked correctly but you you spend less time looking at the code of course you know things like autocomplete and looking at files that's still an amazing experience inside of the anti-gravity editor but for a lot of tasks I think we find and our team finds that we we end up operating at this like higher level right if you have if you're putting together a design doc you know a very common thing you want to hand off a feature to another team you you'll often spend sometime in the agent manager just putting together research and maybe asking and iterating with the agent. Put together this this this artifact to describe how this change is going to go. Share that out with your colleagues and use that as the starting point as opposed to what we traditionally have been sitting inside of the editor and kind of clicking through a bunch of different files. Um it helps you ingest all that context, all that information a bit easier.
>> Yeah, it's a bit more intuitive. And uh actually related to that, we actually did get a question from the audience ahead of time from Dev as well. And um Heck no is asking, "Will the editor evolve to look less like a traditional editor?" as AI assisted editors become more of the norm.
>> Yeah, I guess there's a couple definitions here. Traditional editor, AI assisted editors, we actually took a maybe even a more extreme stance and said the agent manager we wouldn't even classify as an editor. There's no way for you to make a text edit inside of the agent manager. And this is by design. I think we want to give people the AI assisted editor, the escape hatch, the advanced, you know, I need to double click and go into a particular file. Um, but really we wanted to to spend more time orchestrating many agents, orchestrating these powerful agents to abstract away the complexity of the files and of the changes. So, this is where this is where we're putting our money. Uh this is where I think a lot of people are are betting and excited about the direction of the industry where the the nature of your your job the nature of a software engineer is going to change from you know standing inside of a a traditional editor typing lines of code to supervising many agents and seeing you know hundreds of lines getting generated every second.
>> Right. Right. Well, I'm excited for that future. Well, the future is here.
>> The future is here. You can already do this.
>> It's only going to get better.
>> Exactly. Um, Okay. So, Andy, what are we looking at here now?
>> Yeah. So in this case uh we've created a heat map of all of the uh countries. So in this case that the more blue it is, the uh the more metals it has. Um and so you can kind of see and get a sense of the gradient of all of the different countries and seeing which ones have a lot more metals versus which ones do not have any. So um yeah, so that's one iteration of it. There's many different ways that we can try to visualize this such as like uh another follow-up that I can imagine is like okay now apply this heat map onto a world map and like visualize that. Um, so it's really up to the the user at this point on like where they want to take it. And I think that's the magic of it is that like um the the colleague that is anti-gravity will work with you and won't just go ahead and go uh go rogue on many things but instead it kind of like constantly checks in and being like hey like how's this looking so far? Do you like this direction? If so, let's continue. If not, like let's course correct. And that way, it really feels like a partnership and you feel really in control of where you want to drive uh the work, whether it is the topic you want to learn or whether it's this new website that you want to build for the weekend.
>> I feel like now with this new way of like building things, it's like how do you know when you're done or are you ever done? I mean, it feels like you're just constantly just you can iterate so quickly, so it's always a work in progress, but it's also exciting.
>> It is. I don't know if it's great for perfectionists, though. Just kidding.
>> Uh, maybe not.
>> No, but it I think it's a really exciting thing. Um, I think I do kind of want to I mean, first of all, is there anything else you want to share with this because I have a big question.
>> Um, what do you think? I'm excited to hear this big question.
>> Well, I think let's excited.
>> What is the meaning of life? No, I'm just kidding. Um, so I guess like what's your what's your takeaway for how we build things today using anti-gravity? Like it's not just software engineers that are using it. And you talked about thinking, but I mean, it's not just coding tasks, right?
>> Right. It's it's not just coding tasks. Uh so much of it is like um it really ties down to like thinking and like a lot of it in my head is about like being on the like the the same wavelength as uh the other side. And I think that if you meet someone, you want to you usually take a bit of time calibrating to figure out your wavelength, but the more you work with them, the more they understand you. And like sometimes like 30 minutes to into a conversation, you realize you're just on the same wavelength and ideas just start coming. Um, and I think that's one of the most exciting things about this type of collaboration model is that like uh it just takes a bit of time to identify that wavelength. Um, and once you really are on that same wavelength and there's a lot of this this context that you've talked about, then the ideas just start coming and then and I think that at that point in time when you are on the same wavelength, that's when you really start surprising yourself. you're just like, I didn't think of that idea, but like this agent brought all these ideas out of me. And I'm like, you kind of get impressed with yourself. And it's it's almost an addictive feeling of realize that realizing that like um you're able to uh surprise yourself with a lot of these ideas, whether it's this app that you never thought you could build or like I'm pretty sure there's many times when I was like in school, I would be like, "Oh, like it'd be really cool if I can build this." But you're you're trying to visualize the path to getting there. there's all these difficult difficult like concepts and you're like I don't think I can like I don't see the path there and then suddenly with the the with anti-gravity I think like oh like I would love to build this but also like walk me through how you would build this and I want to understand the different pieces as opposed to just building this I want to be a part of the process I don't want to just see the outcome so it ends up being a really fun journey both the process and also in the outcome and it always wants me it always brings me back to square one where I'm like I want restart this and try to build something new um over the weekend or in the evenings.
>> So if you are just trying to solve a problem, it doesn't even have to be a traditional like software development problem. Consider anti-gravity because if you're just using it for coding tests, you're underutilizing it.
>> Totally. Yeah. I think that the the takeaway for me is that like I've obviously everybody only has 24 hours unless you're uh some sort of super super human being u per day. And I think that in the time that I have, I feel like previously I spent a lot of time on on uh Google Chrome where I would browse and like research things, uh search for things. But I've found that more like increasingly I've been using um I've been using anti-gravity a lot more and
It makes me realize that, um, I've shifted a lot of my time from Google Chrome into Anti-Gravity, mainly because I feel like within Anti-Gravity, it has the tools to fetch a lot of information through Chrome. And obviously, I still use Chrome. Chrome is really, really great for doing specific types of research or testing out web apps. But I've really enjoyed using, um, the agent manager within Anti-Gravity to do research, get work done, um, iterate on things, review, uh, work, um, and also collaborating with all my co-workers.
An interesting rule of thumb that I, I try to keep in my head is if I do something more than, let's just say, three times in a week, that is something that the agent should do. And that way of thinking, not necessarily with code. Sure, there are going to be repetitive coding tasks, but I think people are already in the frame of mind to, to sort of use agents for this. But even, um, I was talking to Paige a few weeks ago and she, you know, we were talking about flight checking and like travel planning and and these sorts of things. If you get in the mentality that an agent is very good at boiling the ocean, it's really good at going out, seeking information, bringing it down, and condensing it into something you can understand. That mentality will take you very far for not just coding tasks, but really anything in your life that involves this sort of synthesis of information.
Um, another example is, you know, I'm planning a trip and not only are there logistical considerations with this trip, but as a photographer, I want to see what are the places I'm going to, what are some example photos that you can get at each of these locations. That's not going to be very well summarized inside of a traditional, you know, Gemini app interface. So, what I do instead is, okay, agent manager very good at making visual artifacts. So go out, make me a website, pull in all those different images, make that travel itinerary, but make it an interactive website, right? And this way of thinking where suddenly this previous task that you might have not considered to be part of the scope of what the agent is able to do, if you let your mind be a little bit more flexible, it can really do a lot outside of the scope of just coding.
Yeah. I, I just visualize a future where the modes of how we interact with the online world are going to change. Totally. Like they're going to be very personalized, very customized, whether it be from yourself building it or like the tools that we use are just serving you extremely customized experiences. Totally, like that. So, it's, it's exciting.
Um, I, I think you bring up also another good topic, which is how can you think about using Anti-Gravity versus using, like you said, Chrome or Gemini app. I think a lot of people might be like, well, you know, I use Gemini app, it's more familiar to me, you know, how do you think about those two?
Yeah. Yeah. I think in the short term, you know, what you should consider is Anti-Gravity has access to your computer and your file system in a, in a safe and and and sandboxed way. Um, but what this means is, let's just say you're doing, uh, sort of interactive like data science type of activity or something that requires maybe a CSV that sits on top of your desktop or a file that you just downloaded. Um, this is the type of thing that I would probably lean on Anti-Gravity to do a bit more. Um, anything code adjacent, I think, makes sense to do inside of Anti-Gravity. But, you know, the Gemini app is great for a variety of other use cases. You know, you're on the go, it's mobile. There's a bunch of different strengths. And I think the nice part about frankly what we're doing at Google, which I find really, really exciting is that these products are all, they're unified in the sense that, you know, they're using the Gemini model. All the, when the model gets better, all these products will get better, and we're all contributing to that and making sure that you get the best possible experience across all the different Google products. So in many ways, there are, there are probably more overlaps than there are differences when it comes to using these products.
Yeah. And I think that one thing I want to add on top of that is this is related to like the design philosophy around the agent manager. Um, so historically, you know, we've, uh, there's also the the, um, editor feature, but I think that, uh, when folks currently install, uh, Anti-Gravity, that's that's the first place that they, um, get plopped in. But, uh, increasingly, I, I think the agent manager is the starting entry point. Um, and you can see from the editor, or sorry, not from the editor, but the, the agent manager over here, that there's, there's actually very little that is technical that is on this surface of the screen. Um, it is very much designed to be approachable by anybody. Um, you don't need to have a CS degree to understand what is going on in the screen. And I think that that's kind of tying back to the earlier, uh, concepts that we were talking about, which is like, what does an agent-first experience generally look like, and also how do you provide it with like all the tools that, you know, uh, Kevin was mentioning with like having the sandboxed file system. Um, so I think that having something that is extremely powerful but very approachable for everyone, but also has the capability to go that last mile, like in this case, on the top right, there's this open editor feature if I want to do that, those finishing touches. Um, so the way I think about it is, agent manager can do like 80 to 90% of it, and it also has the extensions needed so that you can wrap up and do that extra 5 to 10% um, if you hop out into the editor. Um, and with that in mind, that basically opens up the doors for the agent manager to everybody because suddenly there, you don't need to have this knowledge or that prerequisite knowledge to use this. Um, you can just go ahead and start using this, similar to how you use a Gemini app. Um, the main upside, as Kevin mentioned, is that there's more tooling, um, and all there's also the agent manager capabilities in a specific aspect, given that it's a local environment. Um, and with that, you can start using it as like the entry point for a lot of your work.
Yeah. Yeah. And you know, there are a lot of tools at our disposal. So I think it's just like really exposing yourself, like just trying it out and seeing what the limitations or possibilities are and what is more useful to what you're trying to achieve. Um, I think, uh, actually, we are getting some questions here on YouTube. Um, there, this person is actually wondering, um, why is Google AI Studio different than Anti-Gravity? Are the models the same on the back end? So I think there's some, yeah, similar questions like how can we think about AI Studio and Anti-Gravity as well.
Totally. Yeah. The AI Studio team, we work very closely together, actually. And and a lot of the underlying capabilities of the agent, you can expect to be very similar, not only because the model is the same, um, but also because the, the way we've constructed the agent is sort of this unified experience across the two products. And what we'd really love to strive for, and we're getting there closer and closer. We're shipping a bunch of great features here with with Logan's team, making sure that AI Studio and Anti-Gravity can play very well together. So what that means is, you know, AI Studio is an amazing way for people to build really quickly a first iteration of what they've dreamed of. There will reach a point, and maybe this is the topic of this, this podcast in general. There'll reach a point when perhaps you need to get a little bit deeper into the code. Maybe you want to bring it onto your local machine. And that's really where the handoff between AI Studio and Anti-Gravity will come into play. Um, there are different sets of features, sure, but I think the biggest thing to think about is you're going to be iterating in very quick succession in in a web environment on a website. Um, if you want to do different types of projects, or if you want to actually get into the file system itself, this is when you'll sort of bring it onto your computer and use Anti-Gravity.
Okay. Yeah, I think that's a great way to think about it. Um, one of the things that we've touched on in the first section here is also just like the interaction modes you have available to you in Anti-Gravity, which I think are a big differentiator for it. So, how can you think about, I mean, we've talked about how it's, you, we're making it more intuitive for people. Um, can you just tell us about how you think about that? Like, why are we so focused on interaction modes with Anti-Gravity?
Um, so for interaction modes, the the most common mode that comes to mind today is the chat box. And I think that historically, um, most folks, most people, all know how to interact via chat. You know, like you use, uh, iMessage to text your, uh, text your family or friends, or you might use Messenger or any other messaging app. Um, and so that was like a pattern that people were already familiar with. And so when we were thinking about building Anti-Gravity, the thing is, like, how can we remove as much learning as possible so that you can just go ahead and it's like you'd already know how to use it. So I think that was the first thing. Um, but when we want to go, when we want to think about going beyond that, the question comes to mind, how do you empower the agent to do even more? And so at that point, the question becomes, is the chat box enough? Um, and so similar to how we were talking about working with a colleague, like one way you talk with a colleague is you just, um, tell them, "Hey, how's it going?" Um, but there are many other ways that you can communicate with them. Uh, whether it's through body language, through gifts, through like all these different ways of communicating. Like we're not limited purely with words. And so, uh, when we were thinking about interaction modes, uh, many of the, the, uh, subsequent ideas is around how can you be more specific with the, the context that you want to provide? How do you specify like the area? How do you specify like what is the, um, which part of the image you want to comment on? How do you, how do you give context by by interacting with the editor? Like, so one of the really cool things that we do within the editor is that we understand which files you're generally looking at, like what you are doing in your terminal. So that when you have a question inside of the agent, it already knows what you were trying to do in the terminal. So as an example, if I was trying to run a like a, a git command and I, I accidentally forgot, if I asked the agent, like it knows that like what I tried, and it, it kind of can infer, oh, like this is what you're trying to do. And so there's many of these different types of signals, uh, whether it's implicit or explicit, to that the agent can gather, uh, so that it can provide more value to you.
I, I'll add one thing, which is just ultimately, we need to build a product around agents. Agents are going to do more. The way that people interact with it is going to change. And this is one of the most important things that we can do as a team to enable people to use agents. It's not as agents are going to do more work, for example, what are the touch points that you as a user will want? Is it going to be every, you know, if you're only interacting with something every 10 minutes versus every 30 seconds, the interaction pattern is going to be very different. And so we focus a lot on interaction because that is really the thing that is shifting underneath our feet. These agents are able to do more, but we need to enable our users to actually have control and and work well with these agents.
Yeah, agreed. And I think there's this shift, right? I, as you mentioned, you have competitive repetitive tasks throughout the week and you're like, okay, now I'm shifting my mindset to saying, oh, let me think about how I can enable an agent to do that instead. But that's already a, I think, a something that people have to get used to, right? Um, and so it seems like you want to make it as easy as possible for that transition to happen. And so interaction modes is like, okay, how would a human interact and say, let me hit up my buddy, can they? Like, if you had an assistant or something, you'd be like, oh, maybe they can take it on. How do we get that mind shift to happen with software instead?
Um, but it's interesting because I feel like the way that we, as you said, the way that we've started with AI applications is through text and chat. So we've actually become so accustomed to that, even in the span of like one year, that now people have to almost like relearn or refocus on like, actually, no, we got, we can do it the way that humans would. So I find it kind of interesting. It's like we started with human interaction, we moved towards browser and mouse and key, and then we go to chatbots, and now we're going back to like that intuitive human interaction mode.
Yeah.
So it's a, I don't know, it's an interesting shift.
It's so interesting as a challenge. Like the other thing that kind of came to mind is that like, like if we kind of zoom back to like, uh, maybe like five, 10 years ago, before agents were even a thing, like most of our interactions were all like at the editor level. So it kind of felt like as if like depending on your level of abstraction, like how you would interact evolves. So initially, like the level of abstraction was just like modifying code directly, and, um, you can only interact something that you can understand. And so I think that that kind of alienated a lot of people who didn't understand. And so as you kind of like zoom out of the abstraction, like one level higher and higher, opens up the doors for a lot of folks. And so, um, obviously, if we, if we work with the constraints that the interactions are always approachable, and that as abstraction gets higher and higher, um, it becomes also like conceptually is also approachable, that suddenly you have something that is usable and also approachable by everybody. And I think that would be an amazing world to have where, like, the Anti-Gravity can be used by everybody to do whatever they want.
Yeah. One of the interesting conversations I've seen happening online from folks on Twitter and people at Google, it's like, you know, we've, we've seen these amazing capabilities with AI software development tools, but they've actually maybe helped more experienced software developers more than junior, because you can actually get that very last 10% where you're like, okay, I got to debug. I can actually go and understand exactly what happened and I have that perspective. Um, I think that's changing over time, of course. What's your take on it in terms of enabling just everyone with higher levels of abstraction, more capabilities, more advanced, like ability to debug to that last 10%?
Yeah, it's tough because what the agent is good at is getting it. It's very different month over month, you know, quarter over quarter. I, I think is is actually interesting, like talking to people about junior engineers who are thinking about what their career should be and whatnot. I, I actually do think these these tools can make it such that everyone sort of gets a huge, a huge lift, and it's a question of who embraces it, almost. If you are just starting out, suddenly your baseline for what is considered an effective engineer is is so much higher, and that that can be very daunting. But at the same time, that that also means you, if you embrace these tools, can become that much more, um, empowered. And so there are different ways, you know, they might harness the fact that you can get from zero to 90 as a junior engineer, but a senior engineer, like you said, will take that last 5% and and really know what to do with that last 5%. So there, the agent has a way of like meeting you sort of where you are. And I like to think that at all experience levels, from someone who's never touched code to someone who's, you know, been soaking in the industry for for 20 plus years, the important thing is that you buy in, that you embrace, and that you are willing to to learn new tools. And I think that, you know, even like the way that we were coding a year ago, in many ways, because things are changing so quickly, it is almost an even playing field where a year ago, the skills and tendencies of reaching for AI are so different than what they are now. And the person who is able to embrace that is going to be competing at a much higher level.
Right. Yeah. And and I can guarantee you that anyone that's watching this, I think is in that bucket of people who are embracing it wholeheartedly.
Totally.
Um, so it's really exciting to see. Um, I do want to bring in a couple more questions. We have a little bit of time left. So would love to see what folks are saying online. Um, let's see. Okay. I'm going to double check what we have. So we have a question here from LinkedIn from Adriana, who asks, do you support versioning or rollback? Can we go back if a generated change doesn't work as expected?
We provide a lot of optionality within, uh, Anti-Gravity. So obviously, it's, uh, we have a built-in IDE, and so there's the version control tools that you love. You can use Git and you can, uh, create commits. You can revert that, and so all that is still built in. Um, why, why don't you just show them? You can, there's a button on each message, um, you go into the conversation, it's this thing called revert. And basically, if you go up to your previous message, say, I, I actually don't like what you just did, just click that button and you'll, you'll be able to revert up to that point. So, you don't have to click it. We kind of like what you did here, but you see the idea.
Yeah. So in this case, um, I can go ahead, clicking revert per message. And so basically, the, the way that the agent captures work is in between the messages, like, okay, what are the changes I, in between, uh, like since that message, and then it creates a diff of all of those, and then it allows you to revert it. So that way, if you want to course correct from that point, because you felt like that was the point where you had a bit of a checkpoint, you're like, okay, I want to, um, maybe fork off of that and start, uh, adding something off of that. Um, I can go ahead and click over here, undo changes up to this point. And upon clicking on it, also shows a confirmation of like, hey, which files are going to be affected. Um, and then you can also view the diff for each of these. And so if I click on confirm, um, it'll go ahead and undo, um, all of my work. And so these are just different ways that you can course correct, whether it's moving forward and adding new changes, or whether you want to move backwards, which is like reverting some of your changes.
Yeah. Yeah. And that's really important, I think, just to give people like that peace of mind, is like, oh, actually, I don't like that. Let's go back. Yeah.
Exactly.
Um, nice. Okay. So, uh, we have another question here from YouTube, and they ask, how can you import projects from VS Code to Anti-Gravity? Can you do that?
Yes, you can. Um, it's a very similar model. If you go back to that original screen, um, you can basically just, uh, what is it? Open workspace and and open that same folder that you would normally open in VS Code, and it'll just appear.
Nice. Very easy. Straightforward. Okay, one more question here from YouTube. Um, what other models can be added to Anti-Gravity? I know we talk a lot about Gemini Flash, which is awesome, but can you use other models?
Yeah. Yeah, we support models from OpenAI. We support models from Anthropic. Um, and we support models from Google. I think there, there's a lot of things that we could talk about in this world. I think with Google specifically, we've built and optimized our product for Gemini, and you can expect the best possible product, you know, for for the Gemini models to be Anti-Gravity and and vice versa. Um, but we do support a lot. We want that sort of open ecosystem style. Uh, we want people to be able to try different models. So, you know, we don't restrict on that.
Yeah, that's great. Um, uh, I'll, I'll squeeze in one more here, uh, from LinkedIn as well. So, as we move into agent-based workflows, um, is Anti-Gravity thinking about how to handle specifically permissions, access control, and boundaries for what each agent is allowed to do?
Totally. I'd say this is one of our our P zeros, if you want to use corporate speak, one of our highest priorities. Um, I think safety is is super, super important for getting people to buy in and trust that these agents won't go off and, you know, do something potentially harmful to your codebase or to your machine. Um, and so we've been investing a lot of time to to make sure that this is done correctly and done well. What I will say is currently we have a number of guardrails and Anti-Gravity is used by a number of teams. Enterprise is, I mean, it's used, we have a different variant internally at at Google, but, you know, a lot of the same principles apply. And so obviously security is is not something that you can just ignore. Um, and so we have things like, uh, domain allow list, for example, in the browser, you can specify which domains you actually want the agent to have access to. You can specify which terminal commands you want it to run. Uh, by default, it will ask you before executing any of these terminal commands. We also have, um, a sandbox mode to essentially make sure that it won't go off and and make harmful changes to files that are outside of your workspace. Your workspace is defined as the project you're working in. If it wanted to go off and access your downloads folder, it won't be able to do that. So, we've done a number of things in this way to just make sure that people feel safe with the app. And I can, you know, personally attest that it, it does do that.
Great. Um, before I, I know we have a couple minutes left, but I did want to touch on just like personal tips, tricks, and like workflows that you use. What are your favorite ways to use Anti-Gravity in your personal workflows? Now, we've talked about, well, before this session, we've talked about planning mode or learning mode. What are those things?
So a lot of my time these days, I haven't been able to get as deep in the weeds for code. Um, and I think I'm thinking a lot more about like, what are the next, uh, cool features we should do? Like, how can we increase, uh, the trust in the agent? And, uh, I usually work very closely with the engineers, and sometimes it's, it's important to be on the same wavelength, uh, as I mentioned earlier. And so what I like to do is I look at the codebase and I want to, uh, I ask, I frequently ask the agent, like, hey, can you explain this? Okay, what, what does this relate to? And typically, like, if I did this five, 10 years ago, I'd be going through every different file and trying to connect the dots. And I think, uh, Anti-Gravity is so good at connecting the dots for me, such that like I can understand things at the right abstraction level, so that I can prepare for different conversations with whoever I'm working with. And so, um, it goes back to using it as a tool for thinking and understanding which abstraction level I should be understanding, um, getting that knowledge from Anti-Gravity, um, and then using that for the purposes of my work.
Nice. How about you?
I'm probably spending, I think my favorite way to use it is is to spend most of my time in the implementation plan, especially with a codebase that is as large as Google's. I mean, you could read about this. It's probably the biggest codebase to ever exist. Um, a lot of work I, I spend most of my time, let's just call it 80% of my time, actually iterating on the plan. And if the plan is pointing to the right files and high level has the right brush strokes, I have much more confidence that the output will be correct. I think if you did not have that, it, it much higher likelihood for going off the rails and maybe changing files, and then at that point, you're on a slippery slope where, you know, you might have to throw away all the work. So I spend a lot of time iterating with the agent before it's even made any changes. And I will do this for three or four different things that I think I'm going to do that day, and then I will fire off those four tasks either in parallel or in serial. That part matters less, more just I'm spending a lot of time planning. And once it has a good plan, I feel pretty confident that it'll be able to do the right thing. I feel like that's also a shift, like, you know, of course, we want to see the magic right away, we wanted to see it build. But are people spending enough time with the foundation and the thinking and thinking about the problem? What is success look like? What is the plan? If I were to think about it first, what do you think?
Yeah, I think it's really, it's a really important stage, which is why our default mode, uh, is planning, because we want to make sure that this is at the forefront of what you should be thinking about. And I think a really good colleague will not only do what you ask it to do, but it also tries to guide you through whatever is the best practice. And so our agent will usually will decide whether a task needs planning or not. And if it does deserve a lot more thinking at that point, it'll ask you like, hey, like, let's collaborate on this. And it'll show you warnings on like, hey, this is actually an important thing to review. Make sure you review this before, uh, proceeding. And so that way, you feel like not only, not only can you trust it for building, but you can also trust it to tell you when you should be thinking more versus less, especially if it's like a humongous refactor. It's very hard, uh, as a challenge to even estimate how much work something can be, whether it's just a one-line change, or whether it's maybe not possible, or whether it requires a significant refactor. And so, uh, with the understanding the agent manager has, it will help you figure out how much thinking you need to do. So that way you understand the work that is involved in whether you want to move forward or you want to change how you approach it.
Yeah. And I think, like, to wrap all of this together, going back to your original point of this is a tool that anyone can use because anyone is a thinker. Anyone has a problem that they can approach and solve and find success for. So if you are a thinker that are trying to, and you're tackling any kind of issue, >> start with the problem. Use Anti-Gravity to iterate on that problem first, and then see what it can build and how it can do that.
Exactly.
Amazing. Well, we are just out of time. So, I just want to thank you both for the amazing conversation and for showing us the agent manager artifacts and interaction modes.
Thanks for having us. Always good to be here.
Yes. Well, thank you everyone for joining our live stream once again. If you want to check out all of the Anti-Gravity resources, check out the description. Uh, just a couple of quick reminders. We just launched GEAR, which is Gemini Enterprise Agent Ready. So, if you are interested in AI agents, they are the biggest platform shift in a decade. GEAR is your new home base to build and scale them from hello world to production. You can get learning paths, earn official badges, connect with a community of builders. Speaking of connecting, mark your calendars. Google Cloud Next 2026 is happening April 22nd through the 24th in Las Vegas, and registration is live right now. We have an incredible lineup of, uh, people that you can meet, and wherever you are in your developer journey, check out all the content. Make sure to check out the registration link and grab your seat before, uh, tickets run out. And we can't wait to see you there. And we'll see you next week on Google Cloud Live. See you everyone. Hey, hey, hey. Hey.