Agentic Sites: Building Hyper Personalized Websites — Carlos Sanchez, Adobe

>> Hello. >> Hello. Thank you for coming. Um I'm going to Thank you for coming. Um I'm going to Thank you for coming. Um I'm going to talk to you about Agility Sites, how we talk to you about Agility Sites, how we talk to you about Agility Sites, how we call it as building hyper-personalized call it as building hyper-personalized call it as building hyper-personalized websites. I'm not going to just talk websites. I'm not going to just talk websites. I'm not going to just talk about it. I'm going to show you what about it. I'm going to show you what about it. I'm going to show you what we're building. we're building. we're building. Um I've been working on on this project Um I've been working on on this project Um I've been working on on this project for for a bit now, and for for a bit now, and for for a bit now, and we'll try to show you what is possible we'll try to show you what is possible we'll try to show you what is possible today with with AI. today with with AI. today with with AI. Uh I work at Adobe at a Uh I work at Adobe at a Uh I work at Adobe at a I'm a principal scientist at a product I'm a principal scientist at a product I'm a principal scientist at a product that not many people know, Adobe that not many people know, Adobe that not many people know, Adobe Experience Manager, content management. Experience Manager, content management. Experience Manager, content management. We run a lot of We run a lot of We run a lot of uh uh uh website properties for big brands, and website properties for big brands, and website properties for big brands, and my background is in in open source, uh my background is in in open source, uh my background is in in open source, uh contributing to to a lot of foundations contributing to to a lot of foundations contributing to to a lot of foundations and projects. and projects. and projects. What are Agility Sites, and how are we What are Agility Sites, and how are we What are Agility Sites, and how are we building this thing? building this thing? building this thing? So, we're looking for sites So, we're looking for sites So, we're looking for sites that are that are that are uh uh uh looking at the what intent the user looking at the what intent the user looking at the what intent the user browsing browsing browsing uh has. What is the user doing? What is uh has. What is the user doing? What is uh has. What is the user doing? What is the user trying to achieve? And the end the user trying to achieve? And the end the user trying to achieve? And the end goal is to personalize these pages for goal is to personalize these pages for goal is to personalize these pages for the for the current user browsing, so the for the current user browsing, so the for the current user browsing, so that eventually this that eventually this that eventually this uh drives uh higher engagement or uh uh drives uh higher engagement or uh uh drives uh higher engagement or uh conversions, whatever the marketing conversions, whatever the marketing conversions, whatever the marketing teams want to want to achieve.

teams want to want to achieve. teams want to want to achieve. And these pages are personalized in real And these pages are personalized in real And these pages are personalized in real time based on the on the user that is time based on the on the user that is time based on the on the user that is uh accessing the site, and what is the uh accessing the site, and what is the uh accessing the site, and what is the what is the user doing. what is the user doing. what is the user doing. The stack we're using is AMH delivery. The stack we're using is AMH delivery. The stack we're using is AMH delivery. So, this is the part of the product we So, this is the part of the product we So, this is the part of the product we we have, we have, we have, uh where all the content is on the edge uh where all the content is on the edge uh where all the content is on the edge and then we have back end service that and then we have back end service that and then we have back end service that powers this experience with powers this experience with powers this experience with different LLM providers LLM services different LLM providers LLM services different LLM providers LLM services we use Cerebras for fast inference or we we use Cerebras for fast inference or we we use Cerebras for fast inference or we can use also we tried bedrock and and a can use also we tried bedrock and and a can use also we tried bedrock and and a bunch of others. I'll be showing bunch of others. I'll be showing bunch of others. I'll be showing Cerebras today Cerebras today Cerebras today and you will see the reason why. and you will see the reason why. and you will see the reason why. The The The the engine that is personalizing this the engine that is personalizing this the engine that is personalizing this this bits is this bits is this bits is using the rich content and blocks. using the rich content and blocks. using the rich content and blocks. So different blocks on the site are So different blocks on the site are So different blocks on the site are customized depending on on what the user customized depending on on what the user customized depending on on what the user persona is.

persona is. persona is. We don't want the the whole site to be We don't want the the whole site to be We don't want the the whole site to be generated. I mean if you talk to generated. I mean if you talk to generated. I mean if you talk to marketing people they they have a very marketing people they they have a very marketing people they they have a very strict brand guidelines. You don't want strict brand guidelines. You don't want strict brand guidelines. You don't want to just to just to just come up with our have some come up with our have some come up with our have some hallucinations there. So the what is hallucinations there. So the what is hallucinations there. So the what is personalized is different sections of personalized is different sections of personalized is different sections of the site and we use the whole site as the site and we use the whole site as the site and we use the whole site as a corpus. We built a rack from the whole a corpus. We built a rack from the whole a corpus. We built a rack from the whole site. So what is generated is grounded site. So what is generated is grounded site. So what is generated is grounded on on the existing site. on on the existing site. on on the existing site. We tried to solve the problem where one We tried to solve the problem where one We tried to solve the problem where one size fits all. We want size fits all. We want size fits all. We want hyper-personalized experiences. Also we hyper-personalized experiences. Also we hyper-personalized experiences. Also we want to help want to help want to help our customers to do more automatic our customers to do more automatic our customers to do more automatic authoring. So not having to create authoring. So not having to create authoring. So not having to create thousands of different variations of the thousands of different variations of the thousands of different variations of the site but use AI for this site but use AI for this site but use AI for this and then do these multiple layers of of and then do these multiple layers of of and then do these multiple layers of of personalization. personalization. personalization. Some examples of what we're doing or Some examples of what we're doing or Some examples of what we're doing or I'll show in the demo. It's I'll show in the demo. It's I'll show in the demo. It's instant persona adaptation, query instant persona adaptation, query instant persona adaptation, query generation when the user search for generation when the user search for generation when the user search for something on the site, the page with the something on the site, the page with the something on the site, the page with the results is customized for them and also results is customized for them and also results is customized for them and also uh, something like recommendations where uh, something like recommendations where uh, something like recommendations where after you browse the site for a period after you browse the site for a period after you browse the site for a period of time, we of time, we of time, we we can create a page that recommends we can create a page that recommends we can create a page that recommends something based on on on what you are something based on on on what you are something based on on on what you are what we think you are looking for.

what we think you are looking for. what we think you are looking for. For marketers, uh, they can define this For marketers, uh, they can define this For marketers, uh, they can define this strategy on natural language, and they strategy on natural language, and they strategy on natural language, and they can use analytics to can use analytics to can use analytics to to drive the loop of personalization, to drive the loop of personalization, to drive the loop of personalization, and what is the end goal, and how this and what is the end goal, and how this and what is the end goal, and how this goes back again to change to adapt the goes back again to change to adapt the goes back again to change to adapt the personalization to improve that uh, personalization to improve that uh, personalization to improve that uh, whole cycle. Everybody's talking about whole cycle. Everybody's talking about whole cycle. Everybody's talking about loops in this conference, so that's loops in this conference, so that's loops in this conference, so that's that's one of the loops there. How the architecture look like? So, it's How the architecture look like? So, it's a dynamic front end with some blocks, a dynamic front end with some blocks, a dynamic front end with some blocks, what I mentioned before, and with uh, what I mentioned before, and with uh, what I mentioned before, and with uh, edge delivery services is basically you edge delivery services is basically you edge delivery services is basically you compose these blocks, and uh, they are compose these blocks, and uh, they are compose these blocks, and uh, they are updated on on real time through with the updated on on real time through with the updated on on real time through with the AI. AI. AI. The back end, The back end, The back end, uh, we uh, we uh, we we do the um, we do the um, we do the um, evaluation of the models and the evaluation of the models and the evaluation of the models and the providers, providers, providers, and one thing we realized is is that and one thing we realized is is that and one thing we realized is is that this is very dependent on the site. So, this is very dependent on the site. So, this is very dependent on the site. So, we have a bunch of prompts, and we look we have a bunch of prompts, and we look we have a bunch of prompts, and we look uh, we run it across a huge variety of uh, we run it across a huge variety of uh, we run it across a huge variety of uh, models and providers, and then we uh, models and providers, and then we uh, models and providers, and then we look at the accuracy, we look at the look at the accuracy, we look at the look at the accuracy, we look at the speed, but this is going to depend speed, but this is going to depend speed, but this is going to depend highly on what type of site, like how highly on what type of site, like how highly on what type of site, like how big is the site, how I don't know, what big is the site, how I don't know, what big is the site, how I don't know, what different different different what different um, what different um, what different um, area is the site targeting, what what area is the site targeting, what what area is the site targeting, what what type of commerce it is, and so on. So, type of commerce it is, and so on. So, type of commerce it is, and so on. So, we we run this this we we run this this we we run this this um, evaluation continuously. We use uh, um, evaluation continuously. We use uh, um, evaluation continuously. We use uh, Promptfoo. Uh, anybody heard about

Promptfoo. Uh, anybody heard about Promptfoo. Uh, anybody heard about Promptfoo? Promptfoo? Promptfoo? Okay, some people. Okay, some people. Okay, some people. So, Promptfoo allows you to evaluate So, Promptfoo allows you to evaluate So, Promptfoo allows you to evaluate models um, prompts against my multiple models um, prompts against my multiple models um, prompts against my multiple models, providers, and you can do local models, providers, and you can do local models, providers, and you can do local models and any of the models and any of the models and any of the a bunch of open AI compatible a bunch of open AI compatible a bunch of open AI compatible uh providers and and uh uh providers and and uh uh providers and and uh a lot of them, basically. a lot of them, basically. a lot of them, basically. We look for two things. Why? Accuracy. We look for two things. Why? Accuracy. We look for two things. Why? Accuracy. That's that's typically what people look That's that's typically what people look That's that's typically what people look for, but also we want the speed because for, but also we want the speed because for, but also we want the speed because we don't want the site generation to we don't want the site generation to we don't want the site generation to take take take more than 1 or 2 seconds, more than 1 or 2 seconds, more than 1 or 2 seconds, right? Because people uh right? Because people uh right? Because people uh this is already this is already this is already uh proven that people want the the uh proven that people want the the uh proven that people want the the faster the site, the more conversions it faster the site, the more conversions it faster the site, the more conversions it it generates or the the better the it generates or the the better the it generates or the the better the experience it is for the user. experience it is for the user. experience it is for the user. Um yeah, what I mentioned is different Um yeah, what I mentioned is different Um yeah, what I mentioned is different sites may have different requirements. sites may have different requirements. sites may have different requirements. Uh so, you may have to run this uh Uh so, you may have to run this uh Uh so, you may have to run this uh evaluation of models depending on the evaluation of models depending on the evaluation of models depending on the site.

site. site. This is a an ex This is a an ex This is a an ex uh we we secured this uh we we secured this uh we we secured this some of these queries, so we have a 15 some of these queries, so we have a 15 some of these queries, so we have a 15 prompts for this example site. prompts for this example site. prompts for this example site. Um we have uh Um we have uh Um we have uh at the top you can see with Cerebras on at the top you can see with Cerebras on at the top you can see with Cerebras on the Gemma 4 model that was announced the Gemma 4 model that was announced the Gemma 4 model that was announced last last week, last last week, last last week, we can get an average latency of 1.1 we can get an average latency of 1.1 we can get an average latency of 1.1 seconds generating a page. seconds generating a page. seconds generating a page. You you can compare that to the second You you can compare that to the second You you can compare that to the second one, which is 4.6 seconds, right? So, one, which is 4.6 seconds, right? So, one, which is 4.6 seconds, right? So, the difference is huge. the difference is huge. the difference is huge. And that's why uh we use Cerebras for And that's why uh we use Cerebras for And that's why uh we use Cerebras for for this use case. for this use case. for this use case. And uh And uh And uh you can see that different providers, you can see that different providers, you can see that different providers, different models have different different models have different different models have different um um um different speeds. And here is uh let me different speeds. And here is uh let me different speeds. And here is uh let me I can show you the whole I can show you the whole I can show you the whole thing here. Not this one, this one, thing here. Not this one, this one, thing here. Not this one, this one, right? So, at the at the bottom we have right? So, at the at the bottom we have right? So, at the at the bottom we have other other tasks. other other tasks. other other tasks. Sometimes uh maybe some of them may be Sometimes uh maybe some of them may be Sometimes uh maybe some of them may be good.

good. good. They don't need to be perfect, but They don't need to be perfect, but They don't need to be perfect, but they're good enough if they're fast they're good enough if they're fast they're good enough if they're fast enough. So, that's going to be the the enough. So, that's going to be the the enough. So, that's going to be the the kind of kind of kind of decisions that you need to make on decisions that you need to make on decisions that you need to make on whether the model is good enough for whether the model is good enough for whether the model is good enough for your use case or not. your use case or not. your use case or not. Yeah, we're looking Yeah, average 1.1 Yeah, we're looking Yeah, average 1.1 Yeah, we're looking Yeah, average 1.1 seconds. And then the the next ones are seconds. And then the the next ones are seconds. And then the the next ones are going from 4 seconds higher. going from 4 seconds higher. going from 4 seconds higher. And you don't need a huge LLM to do this And you don't need a huge LLM to do this And you don't need a huge LLM to do this sort of work because you are generating sort of work because you are generating sort of work because you are generating text, you are deciding where to put text, you are deciding where to put text, you are deciding where to put blocks and how to organize the website, blocks and how to organize the website, blocks and how to organize the website, you don't need a lots of information for you don't need a lots of information for you don't need a lots of information for that. that. that. So, this browsing and the queries So, this browsing and the queries So, this browsing and the queries uh is are being recorded. So, these are uh is are being recorded. So, these are uh is are being recorded. So, these are the metrics or the the the data we gather from the user, the the data we gather from the user, and this is fed into the LLM to and this is fed into the LLM to and this is fed into the LLM to personalize the site. And then in this personalize the site. And then in this personalize the site. And then in this example, we personalize the hero card, example, we personalize the hero card, example, we personalize the hero card, the products, the blog feeds, and and the products, the blog feeds, and and the products, the blog feeds, and and the navigation based based on the the navigation based based on the the navigation based based on the persona.

persona. persona. Also, what are some of the buttons like Also, what are some of the buttons like Also, what are some of the buttons like our call to action navigation, you can our call to action navigation, you can our call to action navigation, you can also we can also personalize those. also we can also personalize those. also we can also personalize those. We We We we create and I'll show you the a for we create and I'll show you the a for we create and I'll show you the a for you page, which is a recommendation. you page, which is a recommendation. you page, which is a recommendation. And this is a interesting one because And this is a interesting one because And this is a interesting one because this you could this you could this you could pre-generate, right? As the user browses pre-generate, right? As the user browses pre-generate, right? As the user browses your site, you gather these signals, and your site, you gather these signals, and your site, you gather these signals, and you could keep generating in this. So, you could keep generating in this. So, you could keep generating in this. So, in this case, you wouldn't need so such in this case, you wouldn't need so such in this case, you wouldn't need so such a big speed. a big speed. a big speed. But But But but that's interesting because it it but that's interesting because it it but that's interesting because it it would be if a user wanted to would be if a user wanted to would be if a user wanted to buy something, you could just say, buy something, you could just say, buy something, you could just say, "Okay, for you, I will recommend these "Okay, for you, I will recommend these "Okay, for you, I will recommend these three products or or something like three products or or something like three products or or something like that." that." that." Um Um Um Yeah, and then they can see this Yeah, and then they can see this Yeah, and then they can see this recommendation, and if they go there, recommendation, and if they go there, recommendation, and if they go there, that that could be pre-fetched for them. that that could be pre-fetched for them. that that could be pre-fetched for them. And obviously, you have to keep updating And obviously, you have to keep updating And obviously, you have to keep updating it as the user navigates around the site it as the user navigates around the site it as the user navigates around the site and and so on. So that that's also and and so on. So that that's also and and so on. So that that's also something to consider on the cost cost something to consider on the cost cost something to consider on the cost cost implications of doing multiple implications of doing multiple implications of doing multiple generations, multiple LLM calls.

When When when the user runs a query, dynamic when the user runs a query, dynamic when the user runs a query, dynamic personalized page is shown to them. personalized page is shown to them. personalized page is shown to them. When the When the When the these queries are also grouped into these queries are also grouped into these queries are also grouped into personas or intent types. So what is personas or intent types. So what is personas or intent types. So what is this guy what is this guy trying to do this guy what is this guy trying to do this guy what is this guy trying to do in the site? Is trying to buy something? in the site? Is trying to buy something? in the site? Is trying to buy something? Is trying to just get information? So Is trying to just get information? So Is trying to just get information? So you can get marketers to decide what you can get marketers to decide what you can get marketers to decide what type of groups, how many groups you want type of groups, how many groups you want type of groups, how many groups you want to have, how you want to deal with with to have, how you want to deal with with to have, how you want to deal with with customers. And the AI will choose the customers. And the AI will choose the customers. And the AI will choose the the blocks and the suggestions for for the blocks and the suggestions for for the blocks and the suggestions for for those groups of people. those groups of people. those groups of people. Um Um Um And we can adopt yes, the the different And we can adopt yes, the the different And we can adopt yes, the the different blocks, the the the sequence of the blocks, the the the sequence of the blocks, the the the sequence of the blocks and blocks and blocks and media. You could also do media. One of media. You could also do media. One of media. You could also do media. One of the things we consider is the things we consider is the things we consider is there was some a model announced there was some a model announced there was some a model announced today or yesterday the today or yesterday the today or yesterday the the nano banana light. So you could even the nano banana light. So you could even the nano banana light. So you could even generate images generate images generate images very fast on the fly.

very fast on the fly. very fast on the fly. Obviously not as fast as text, but Obviously not as fast as text, but Obviously not as fast as text, but that's also something that would be that's also something that would be that's also something that would be I don't I don't know if it's that I don't I don't know if it's that I don't I don't know if it's that something like marketing people would something like marketing people would something like marketing people would want to have generated images. That want to have generated images. That want to have generated images. That depends on on the quality a lot if it's depends on on the quality a lot if it's depends on on the quality a lot if it's on brand. on brand. on brand. And And And the site in this example we have a a the site in this example we have a a the site in this example we have a a product site product site product site and then we have guides, experiences, and then we have guides, experiences, and then we have guides, experiences, blocks and the whole response of the LLM blocks and the whole response of the LLM blocks and the whole response of the LLM is grounded there. And is grounded there. And is grounded there. And there's comparisons. We can do there's comparisons. We can do there's comparisons. We can do comparisons between products that are comparisons between products that are comparisons between products that are tailor and the product pages can be tailor and the product pages can be tailor and the product pages can be tailored for the for the user. Okay, this is this is a bit of the Okay, this is this is a bit of the stack. stack. stack. Um not going to spend too much time Um not going to spend too much time Um not going to spend too much time here, but the browser you have some here, but the browser you have some here, but the browser you have some layers. You have the browser where the layers. You have the browser where the layers. You have the browser where the signals get signals get signals get get get get uh uh uh from I got I got I got so from I got I got I got so from I got I got I got so I got from the from the user and then we I got from the from the user and then we I got from the from the user and then we have the back end. Uh have the back end. Uh have the back end. Uh we can have the back end. We run this we can have the back end. We run this we can have the back end. We run this some of these in in Google. Some of some of these in in Google. Some of some of these in in Google. Some of these are in our Cloudflare. So, the these are in our Cloudflare. So, the these are in our Cloudflare. So, the back end is basically just calling the back end is basically just calling the back end is basically just calling the LLM and doing some reasoning using the LLM and doing some reasoning using the LLM and doing some reasoning using the rack that is built on on the site to do rack that is built on on the site to do rack that is built on on the site to do the generation.

the generation. the generation. And you have obviously you have to have And you have obviously you have to have And you have obviously you have to have the vector database, the inference uh the vector database, the inference uh the vector database, the inference uh machinery and uh machinery and uh machinery and uh that obvious business manager is doing that obvious business manager is doing that obvious business manager is doing the serving the the serving the the serving the the the the at the edge is serving the the pages and at the edge is serving the the pages and at the edge is serving the the pages and the static content. the static content. the static content. So, let me show you because I think this So, let me show you because I think this So, let me show you because I think this is uh so, we call this uh audience of is uh so, we call this uh audience of is uh so, we call this uh audience of one one one because the idea of in marketing because the idea of in marketing because the idea of in marketing they they always dream on being able to they they always dream on being able to they they always dream on being able to personalize things for each individual. personalize things for each individual. personalize things for each individual. So, we call it yeah audience of one. So, So, we call it yeah audience of one. So, So, we call it yeah audience of one. So, I have this this site. Uh this is a site I have this this site. Uh this is a site I have this this site. Uh this is a site that is absolutely generated that is absolutely generated that is absolutely generated uh example site. It's a coffee uh example site. It's a coffee uh example site. It's a coffee uh machinery. So, I can go and and read uh machinery. So, I can go and and read uh machinery. So, I can go and and read some stories some stories some stories and I can go and look at some products. and I can go and look at some products. and I can go and look at some products. Let's go and look at this product. Let's go and look at this product. Let's go and look at this product. I can spend some time here.

Uh Uh let's go and click let's go and click let's go and click here. here. here. Okay, so I'm I'm browsing around the Okay, so I'm I'm browsing around the Okay, so I'm I'm browsing around the site and I have this debugging tool site and I have this debugging tool site and I have this debugging tool thing uh which thing uh which thing uh which Uh Uh Uh let me go here, I think. Let's see. Let's see. So, down there is the signals that the So, down there is the signals that the So, down there is the signals that the that the browsing that the browsing that the browsing is giving us. So, I don't know if you is giving us. So, I don't know if you is giving us. So, I don't know if you can see it much because I cannot see it can see it much because I cannot see it can see it much because I cannot see it much. The So, much. The So, much. The So, the user is bucketed into the exploring the user is bucketed into the exploring the user is bucketed into the exploring category. We have the pages that have category. We have the pages that have category. We have the pages that have have visited, and then we have how much have visited, and then we have how much have visited, and then we have how much time is spending on each page. All of time is spending on each page. All of time is spending on each page. All of this data is now available for the LLM. this data is now available for the LLM. this data is now available for the LLM. So, So, So, if I go here, I already have a for you if I go here, I already have a for you if I go here, I already have a for you page that was generated for me page that was generated for me page that was generated for me and and and based on my browser.

And you will not notice that it's And you will not notice that it's slightly different than everything else, slightly different than everything else, slightly different than everything else, but if I go here and I run a query but if I go here and I run a query but if I go here and I run a query like I want I'm looking for a coffee like I want I'm looking for a coffee like I want I'm looking for a coffee machine to machine to machine to uh prepare uh prepare uh prepare coffee coffee coffee while camping. The site is this was just generated for The site is this was just generated for me. me. me. And then you're going to see some things And then you're going to see some things And then you're going to see some things like the text is customized. Camping like the text is customized. Camping like the text is customized. Camping shouldn't mean compromising on your shouldn't mean compromising on your shouldn't mean compromising on your uh whatever routine. uh whatever routine. uh whatever routine. Uh the coffee tips for camping um Uh the coffee tips for camping um Uh the coffee tips for camping um machinery that are being recommended are machinery that are being recommended are machinery that are being recommended are coffee agile and um or the nano, which coffee agile and um or the nano, which coffee agile and um or the nano, which are are are good for good for good for for the for the for the for a camping trip, right? for a camping trip, right? for a camping trip, right? So, you saw how fast this was. So, you saw how fast this was. So, you saw how fast this was. I'm going to run it here something I'm going to run it here something I'm going to run it here something similar that I had here and I can run it similar that I had here and I can run it similar that I had here and I can run it on the debug mode here.

on the debug mode here. on the debug mode here. And you will see, let's make this And you will see, let's make this And you will see, let's make this bigger. bigger. bigger. Total time 164 seconds to generate the Total time 164 seconds to generate the Total time 164 seconds to generate the page. So, this includes a round trip to page. So, this includes a round trip to page. So, this includes a round trip to the LLM. This is using Cerebras Gemma 4. the LLM. This is using Cerebras Gemma 4. the LLM. This is using Cerebras Gemma 4. So, the the Gemma model from Google So, the the Gemma model from Google So, the the Gemma model from Google running on Cerebras on their running on Cerebras on their running on Cerebras on their very fast chips. very fast chips. very fast chips. Uh we get Uh we get Uh we get 2,300 2,300 2,300 tokens per second. tokens per second. tokens per second. Which is not bad. Which is not bad. Which is not bad. I would say. I would say. I would say. >> [snorts] >> [snorts] >> [snorts] >> And if I run it again, uh probably >> And if I run it again, uh probably >> And if I run it again, uh probably something like that. something like that. something like that. Uh the LLM time is 1 second. And again, Uh the LLM time is 1 second. And again, Uh the LLM time is 1 second. And again, 2,200 tokens per second. 2,200 tokens per second. 2,200 tokens per second. This is something that we only dreamed This is something that we only dreamed This is something that we only dreamed about before. On the on this site example site, we On the on this site example site, we have some other options. have some other options. have some other options. Uh Uh Uh so, because we we've been showing this so, because we we've been showing this so, because we we've been showing this to customers, so we have the to customers, so we have the to customers, so we have the the ability to change the different the ability to change the different the ability to change the different models, temper temperature, tokens, and models, temper temperature, tokens, and models, temper temperature, tokens, and so on. And we can uh so on. And we can uh so on. And we can uh we can we can we can show uh and try the different models and show uh and try the different models and show uh and try the different models and see how they behave. Besides the see how they behave. Besides the see how they behave. Besides the automatic test with Prompt Full, then we automatic test with Prompt Full, then we automatic test with Prompt Full, then we can uh manually come and and click can uh manually come and and click can uh manually come and and click things and see and see how that how that things and see and see how that how that things and see and see how that how that works.

works. works. And uh And uh And uh we also have we also have we also have uh OfOneLabs. uh OfOneLabs. uh OfOneLabs. So, we have we build this tool that So, we have we build this tool that So, we have we build this tool that generates an agentic site for any site generates an agentic site for any site generates an agentic site for any site we want. So, if somebody wants to have a we want. So, if somebody wants to have a we want. So, if somebody wants to have a demo for a customer, come here demo for a customer, come here demo for a customer, come here and enter the URL. In less than an hour, and enter the URL. In less than an hour, and enter the URL. In less than an hour, you have an agentic site. I did this you have an agentic site. I did this you have an agentic site. I did this last week with the AI engineering site. last week with the AI engineering site. last week with the AI engineering site. And And And I got this site that is just a search I got this site that is just a search I got this site that is just a search box and a few things. box and a few things. box and a few things. And And And let me open it here, the full page. Not let me open it here, the full page. Not let me open it here, the full page. Not this one. Yeah, okay. this one. Yeah, okay. this one. Yeah, okay. So, I could say So, I could say So, I could say Europe AI conferences. Europe AI conferences. Europe AI conferences. So, these suggestions are also AI So, these suggestions are also AI So, these suggestions are also AI generated. And I get a page that is generated. And I get a page that is generated. And I get a page that is more focused on more focused on more focused on It should be more focused on on the on It should be more focused on on the on It should be more focused on on the on this European conferences. this European conferences. this European conferences. If I go back, did I go I can search for anything the same way I I can search for anything the same way I did with with the Arco. So, I as a did with with the Arco. So, I as a did with with the Arco. So, I as a specific There was someone that was specific There was someone that was specific There was someone that was generating a good comparison side to generating a good comparison side to generating a good comparison side to side.

side. side. Let me see if this one. Okay, here. This Let me see if this one. Okay, here. This Let me see if this one. Okay, here. This one. one. one. I went and this generated a page with I went and this generated a page with I went and this generated a page with a pretty good comparison. If I'm looking a pretty good comparison. If I'm looking a pretty good comparison. If I'm looking at two conferences and I need to decide, at two conferences and I need to decide, at two conferences and I need to decide, if I figure out that the user wants to if I figure out that the user wants to if I figure out that the user wants to do that, this is great because that do that, this is great because that do that, this is great because that gives them a side-by-side comparison on gives them a side-by-side comparison on gives them a side-by-side comparison on the fly. the fly. the fly. Now, Now, Now, this this is I think this is cool this this is I think this is cool this this is I think this is cool already, but then we have I have this already, but then we have I have this already, but then we have I have this idea that idea that idea that probably the I'm probably the I'm probably the I'm a bunch of people are we are talking a bunch of people are we are talking a bunch of people are we are talking about is the web that is is the web the about is the web that is is the web the about is the web that is is the web the future still and so on. Nobody knows. future still and so on. Nobody knows. future still and so on. Nobody knows. But we can also do something with this But we can also do something with this But we can also do something with this with this audience of one, this with this audience of one, this with this audience of one, this generative sites. So, imagine you have generative sites. So, imagine you have generative sites. So, imagine you have you have your personal assistant and you you have your personal assistant and you you have your personal assistant and you ask a query through in this case through ask a query through in this case through ask a query through in this case through Google and you say I want to buy I don't Google and you say I want to buy I don't Google and you say I want to buy I don't remember what the query said. It was remember what the query said. It was remember what the query said. It was something like I want to buy a machine something like I want to buy a machine something like I want to buy a machine and I get this on my Google TV.

and I get this on my Google TV. and I get this on my Google TV. Right? So, this is absolutely Right? So, this is absolutely Right? So, this is absolutely personalized to my query. personalized to my query. personalized to my query. Okay? No, go back. This is absolutely personalized to my This is absolutely personalized to my query. So, I'm there in my living room. query. So, I'm there in my living room. query. So, I'm there in my living room. I don't need a phone, I don't need a I don't need a phone, I don't need a I don't need a phone, I don't need a computer, I don't need anything, just my computer, I don't need anything, just my computer, I don't need anything, just my voice and something that will voice and something that will voice and something that will kind of show me kind of show me kind of show me something that is absolutely something that is absolutely something that is absolutely personalized to to me. Okay, so that one. Okay, so that one. So, So, So, what I was trying to show and hopefully what I was trying to show and hopefully what I was trying to show and hopefully you remember from this session is that you remember from this session is that you remember from this session is that this is now possible. this is now possible. this is now possible. It's only going to get better from here It's only going to get better from here It's only going to get better from here on. It's only going to get cheaper, it's on. It's only going to get cheaper, it's on. It's only going to get cheaper, it's only going to get faster. only going to get faster. only going to get faster. And you will uh be able to have uh And you will uh be able to have uh And you will uh be able to have uh huge personalization options for sites huge personalization options for sites huge personalization options for sites and for other things. and for other things. and for other things. And you can do this with intent driven. And you can do this with intent driven. And you can do this with intent driven. So, what is the what is my user trying So, what is the what is my user trying So, what is the what is my user trying to do? What does my user want to buy?

to do? What does my user want to buy? to do? What does my user want to buy? These sort of questions. And you can uh These sort of questions. And you can uh These sort of questions. And you can uh assemble a page just for them. assemble a page just for them. assemble a page just for them. And you can also do this with uh And you can also do this with uh And you can also do this with uh multiple models and and eventually multiple models and and eventually multiple models and and eventually it's just going to be it's just going to be it's just going to be faster and faster, right? faster and faster, right? faster and faster, right? So, So, So, that's it. Um that's it. Um that's it. Um thank you for coming and I hope you you thank you for coming and I hope you you thank you for coming and I hope you you got the idea. Thanks. got the idea. Thanks. got the idea. Thanks. >> [applause]