youtube / video
Real-world AI agents with Vercel's eve (our actual setup)
About this video
A real-world walkthrough of using Vercel's eve to run agents on our own site - updating CMS comparison pages, posting reports to Slack, and orchestrating multiple agents on cron jobs.
Full transcript
Edited transcript from the original Roboto page
Hey folks, how's it going? Hope you're having a great day. So I want to talk about Eve. And if you don't know what Eve is, it's basically a way of building AI agents. Whether it's kind of like on your website, whether it's doing changes, or it might just be a chatbot within Slack or something. Um, I want to give you kind of like the top level overview. I don't often see people show real world examples of these things. They're very nice in a demo environment, but those demo environments are great until you actually use them in the real world. So this is a real world example. This is where we're using Eve on our own website. And something that you might want to think about if you're doing like comparison pages on your own website. This is a great way of doing it. So we have all these different URLs. Um, I'm hoping we don't get in trouble with Google at some point because these do have like a programmatic nature to them, but we try to keep them up to date and fresh with very organic information.
Um, so if we're looking at these pages, we have this last verified date, which is the 17th of July, which luckily is today. Um, so this for example is going to take information called challenges with WordPress and tell you all the challenges that WordPress has and then tell you how good Sanity is because of the URL structure. It's got the problem platform that we're moving away from and the solution platform that we're moving to. Now we generally do a hell of a lot of this kind of work. We do a lot of headless CMS migrations. Um, but we have to try and keep these up to date. Now we use YAML behind the scenes to be able to keep all that information in one particular file. And the beauty of having a file-based system is that you can keep these updated and you can create pull requests to change things inside of these. And so this is a perfect example of something that you want to do with Eve because it's versionable, it's simple, and it's kind of very deterministic. And the reason why I use the word deterministic is because if something is deterministic, it's way easier to do it with AI and it costs way less money on tokens. And we'll come back to that at some point because token spend is going to be a hot topic. It probably already is, but in the next six months it's going to go through the roof. Um, but you can see here this is getting updated.
Now the second part of Eve, and this is kind of like an announcement that came along with Eve, which I would argue is probably the most important announcement, is this connect, this vercel.com/connect. Uh, this is a way to be able to make these agents just integrate directly with whatever tools you're using. Uh, in my particular case, it's Slack and Linear we use for the vast majority of time, but you could use this with GitHub. You could use it with any of these. You can see here just below, you can click any of these and they'll work, which is really neat. Uh, and all that changes here is just throwing this in and it's all connected inside of Vercel.
Now let me show you in Zed um some of the findings in here. So this is basically like an example finding. So we have uh this updater bot that basically goes out, uses sub-agents to find all this information and create a pull request um to update that information. So in this case, it's like, oh, here's the Framer FAQs. Here's some uh official URLs where we can have a quick look at all the different Framer things. Um, and you can basically set these things up. This is inside of the actual agent skills. So we have these agents. We're a little bit, we're sloppy on the naming convention right now. It's going to get better and we'll release some guides as to how to do this. Um, but essentially right now we have like negative features. So no self-hosting. So Framer sites are locked to Framer hosting. Uh, here's the sources where this is coming from because what you don't want to do uh is put all this information in there that's incorrect without the sources, without it being generated. Uh if you have all these sources in there and you're generating this stuff, I actually personally think it's better than a human researched uh system because it knows exactly what it's looking for. It knows the fors and againsts. Imagine what you can do with this with comparison pages, you know, from one software company to another. Everybody does it. Don't get me wrong, it's not the most ethically like outstanding, really really nice way of doing business, but it is something that pretty much everybody does comparison pages.
Um, but yeah, you can see this is an example finding and then you can create skills based on these. Here's like the guard rails. Um, so skeptical voices, all kinds of different things you might want to put, you know, like this is uh the way that we talk, this is our messaging. There's going to be something if you go dig through this. I bet I've showed something here where I shouldn't have showed it, but you get the idea, right? Um, and so what's interesting about this is once you've got all this together, you can create a loop that basically keeps going and keeps going and you can set it on cron jobs. Uh, we set a whole bunch on cron jobs inside of here. Let me just move this aside and find it. Uh, where are we? There is a cron in here somewhere. Schedules. So you can see CMS freshness, for example, is on the cron job. Um, and I can never read these things, but this is on I think once a week and I think it's on a Thursday from memory. Um, but essentially every time it goes and runs through that cron job, you're basically updating across all the different CMSs that information. So you've actually basically created a self-updating bot. And guess what the bot does after it? It tells you this information. And so once you've built one of these things end to end, it then becomes very simple to create the next one, and the next one, and the next one. And you can either do it in two different ways. You can either make them in isolation. So you can have all these different agents that don't really talk to each other or you can do like we've done here where we have, you know, like competitor intelligence. We have daily SEO, we have a Reddit scout, uh we have like route strategies, uh CMS sweeps, all kind of stuff. Like there's a lot of information that's in here. Um and again, a lot of different skills. So I'll show you just a very quick example inside of our Slack. I'm trying my hardest to not accidentally like dox anything or put anything out there, but if you look inside of Slack here, this is a great example of it. So um, inside of here, we actually have a website report bot. And this website report bot takes information from um, Vercel observability and a whole bunch of other different touch points and basically tells us what's going off. And because we've got all this information ahead of us, we can actually make educated uh decisions as to how we progress the website. So this is a really cool feedback loop and it's been pretty much vital for our growth as a business. Um, where you can continuously spin these things up, continuously rely on these. And the best bit of it all is I put like a hundred dollars on our AI uh whatever it is, the AI gateway. Um, and I haven't spent it yet. And we've got like five different um agents that are constantly running. And we've been doing this since pretty much day one of Eve dropping. Um, not Eve dropping, but like Eve the tool dropping. Um, and so it's yeah, it's pretty pretty wild. It's pretty cool.
Um, and I would definitely recommend that this is going to be the way that people use agents more with being able to take all these different disparate systems, pile them into a single Slack channel and not making it overwhelming. Um there is a different way of doing this completely separately as well which is kind of interesting which is you can use things like Claude Schedule but the problem with this is that Claude Schedule and I can talk about this in another uh video if anybody's interested but Claude Schedule is a non-deterministic way of doing this and it's pretty difficult to get it to be deterministic. So it means that you get kind of like instead of rigid things where it tells you and being able to kind of piecemeal get better and better. It's kind of like a scattershot where it might answer this way one week and it might answer this way another week. And the problem is it's very difficult then to you know take that information week by week and compare it to one another if it doesn't match the same data structure.
So I'm pretty heavily, I'm I hate this expression, but I'm pretty bullish on Eve being like a long-term uh tool that we use internally to take all these different systems, mesh them together, and just tell me what the hell's going off on the website rather than being completely clued out and not understanding how, you know, like Google Analytics works, Google Search Console works, uh Post Hog, Ahrefs or whatever, different disparate systems. You get the idea. Anyway, if you want more info about this, drop us a comment down below and we'll uh we'll get back to you. Thanks for watching. Have a great day. Take care. Bye-bye.
Read timestamped source captions
0:02[music]
0:02[music] >> Hey folks, how's it going? Hope you're
0:04>> Hey folks, how's it going? Hope you're
0:04>> Hey folks, how's it going? Hope you're having a great day. So, I want to talk
0:06having a great day. So, I want to talk
0:06having a great day. So, I want to talk about Eve and if you don't know what Eve
0:08about Eve and if you don't know what Eve
0:08about Eve and if you don't know what Eve is, it's basically a way of building AI
0:11is, it's basically a way of building AI
0:11is, it's basically a way of building AI agents uh whether it's kind of like on
0:13agents uh whether it's kind of like on
0:13agents uh whether it's kind of like on your website, whether it's doing
0:15your website, whether it's doing
0:15your website, whether it's doing changes, or it might just be a chatbot
0:17changes, or it might just be a chatbot
0:17changes, or it might just be a chatbot within Slack or something.
0:19within Slack or something.
0:19within Slack or something. Um I want to give you kind of like the
0:20Um I want to give you kind of like the
0:20Um I want to give you kind of like the top-level overview. I don't often see
0:23top-level overview. I don't often see
0:23top-level overview. I don't often see people show real-world examples of these
0:26people show real-world examples of these
0:26people show real-world examples of these things. They're very nice in a demo
0:27things. They're very nice in a demo
0:28things. They're very nice in a demo environment, but those demo environments
0:30environment, but those demo environments
0:30environment, but those demo environments are great until you actually use them in
0:32are great until you actually use them in
0:32are great until you actually use them in a real world. So, um this is a
0:34a real world. So, um this is a
0:34a real world. So, um this is a real-world example. This is where we're
0:36real-world example. This is where we're
0:36real-world example. This is where we're using Eve on our own website and
0:38using Eve on our own website and
0:38using Eve on our own website and something that you might want to think
0:39something that you might want to think
0:39something that you might want to think about if you're doing like comparison
0:41about if you're doing like comparison
0:41about if you're doing like comparison pages on your own website, this is a
0:42pages on your own website, this is a
0:43pages on your own website, this is a great way of doing it.
0:45great way of doing it.
0:45great way of doing it. So, we have all these different URLs. Um
0:48So, we have all these different URLs. Um
0:48So, we have all these different URLs. Um I'm hoping we don't get in trouble with
0:50I'm hoping we don't get in trouble with
0:50I'm hoping we don't get in trouble with Google at some point because these do do
0:52Google at some point because these do do
0:52Google at some point because these do do have like a programmatic nature to them,
0:54have like a programmatic nature to them,
0:54have like a programmatic nature to them, but we try to keep them up-to-date and
0:56but we try to keep them up-to-date and
0:56but we try to keep them up-to-date and fresh with very organic information.
0:59fresh with very organic information.
0:59fresh with very organic information. Um so, if we're looking at these pages,
1:01Um so, if we're looking at these pages,
1:01Um so, if we're looking at these pages, we have this last verified date, which
1:03we have this last verified date, which
1:03we have this last verified date, which is the 17th of July, which luckily is
1:06is the 17th of July, which luckily is
1:06is the 17th of July, which luckily is today. Um so, this for example is going
1:10today. Um so, this for example is going
1:10today. Um so, this for example is going to take information called challenges
1:11to take information called challenges
1:12to take information called challenges with WordPress and tell you all the
1:13with WordPress and tell you all the
1:13with WordPress and tell you all the challenges that WordPress has
1:15challenges that WordPress has
1:15challenges that WordPress has and then tell you how good Sanity is
1:17and then tell you how good Sanity is
1:17and then tell you how good Sanity is because of the URL structure. It's got
1:20because of the URL structure. It's got
1:20because of the URL structure. It's got the problem platform that we're moving
1:22the problem platform that we're moving
1:22the problem platform that we're moving away from and the solution platform that
1:24away from and the solution platform that
1:24away from and the solution platform that we're moving to. Now, we generally do a
1:27we're moving to. Now, we generally do a
1:27we're moving to. Now, we generally do a hell of a lot of this kind of work. We
1:28hell of a lot of this kind of work. We
1:28hell of a lot of this kind of work. We do a lot of headless CMS migrations, um
1:31do a lot of headless CMS migrations, um
1:31do a lot of headless CMS migrations, um but we have to try and keep these
1:32but we have to try and keep these
1:32but we have to try and keep these up-to-date. Now, we use YAML behind the
1:35up-to-date. Now, we use YAML behind the
1:35up-to-date. Now, we use YAML behind the scenes to be able to keep all that
1:37scenes to be able to keep all that
1:37scenes to be able to keep all that information in one particular file and
1:39information in one particular file and
1:39information in one particular file and the beauty of having a file-based system
1:41the beauty of having a file-based system
1:41the beauty of having a file-based system is that you can keep these updated and
1:43is that you can keep these updated and
1:43is that you can keep these updated and you can create pull requests to change
1:45you can create pull requests to change
1:45you can create pull requests to change things inside of these. And so, this is
1:47things inside of these. And so, this is
1:47things inside of these. And so, this is a perfect example of something that you
1:50a perfect example of something that you
1:50a perfect example of something that you want to do with Eve because it's
1:52want to do with Eve because it's
1:52want to do with Eve because it's versionable, it's simple, and and kind
1:55versionable, it's simple, and and kind
1:55versionable, it's simple, and and kind of it's very deterministic. And the
1:57of it's very deterministic. And the
1:57of it's very deterministic. And the reason why I use the word deterministic
1:59reason why I use the word deterministic
1:59reason why I use the word deterministic is because if something is
2:00is because if something is
2:00is because if something is deterministic, it's way easier to do it
2:02deterministic, it's way easier to do it
2:02deterministic, it's way easier to do it with AI and it costs way less money on
2:05with AI and it costs way less money on
2:05with AI and it costs way less money on tokens.
2:06tokens.
2:06tokens. And you'll we'll come back to that at
2:07And you'll we'll come back to that at
2:07And you'll we'll come back to that at some point because token spend is going
2:09some point because token spend is going
2:09some point because token spend is going to be a hot topic.
2:10to be a hot topic.
2:10to be a hot topic. It probably already is, but in the next
2:12It probably already is, but in the next
2:12It probably already is, but in the next 6 months it's going to go through the
2:136 months it's going to go through the
2:136 months it's going to go through the roof. Um but you can see here, this is
2:15roof. Um but you can see here, this is
2:15roof. Um but you can see here, this is getting updated. Now, the second part of
2:18getting updated. Now, the second part of
2:18getting updated. Now, the second part of Eve, and this is kind of like an
2:19Eve, and this is kind of like an
2:20Eve, and this is kind of like an announcement that came along with Eve,
2:21announcement that came along with Eve,
2:21announcement that came along with Eve, which I would argue is probably the most
2:23which I would argue is probably the most
2:23which I would argue is probably the most important announcement, is this connect,
2:25important announcement, is this connect,
2:26important announcement, is this connect, this versel.com/connect.
2:28this versel.com/connect.
2:28this versel.com/connect. Uh this is a way to be able to make
2:30Uh this is a way to be able to make
2:30Uh this is a way to be able to make these agents just integrate directly
2:32these agents just integrate directly
2:32these agents just integrate directly with whatever tools you're using. Uh in
2:34with whatever tools you're using. Uh in
2:34with whatever tools you're using. Uh in my particular case, it's Slack and
2:36my particular case, it's Slack and
2:36my particular case, it's Slack and Linear we use for the vast majority of
2:38Linear we use for the vast majority of
2:38Linear we use for the vast majority of time, but you could use this with
2:39time, but you could use this with
2:39time, but you could use this with GitHub, you could use it with any of
2:41GitHub, you could use it with any of
2:41GitHub, you could use it with any of these. You can see here just below, you
2:42these. You can see here just below, you
2:42these. You can see here just below, you can click any of these and they'll work,
2:45can click any of these and they'll work,
2:45can click any of these and they'll work, which is really neat. Uh and all that
2:46which is really neat. Uh and all that
2:46which is really neat. Uh and all that changes here is just throwing this in
2:48changes here is just throwing this in
2:48changes here is just throwing this in and it's all connected inside of Vercel.
2:51and it's all connected inside of Vercel.
2:52and it's all connected inside of Vercel. Now, let me show you in Z, um some of
2:54Now, let me show you in Z, um some of
2:54Now, let me show you in Z, um some of the findings in here. So, this is this
2:55the findings in here. So, this is this
2:55the findings in here. So, this is this is basically like an example finding.
2:57is basically like an example finding.
2:57is basically like an example finding. So, we have uh this update bot that
3:00So, we have uh this update bot that
3:00So, we have uh this update bot that basically goes out, uses sub-agents to
3:03basically goes out, uses sub-agents to
3:03basically goes out, uses sub-agents to find all this information, and create a
3:05find all this information, and create a
3:05find all this information, and create a pull request um to update that
3:08pull request um to update that
3:08pull request um to update that information. So, in this case it's like,
3:09information. So, in this case it's like,
3:09information. So, in this case it's like, "Oh, here's the Framer FAQs. Here's some
3:12"Oh, here's the Framer FAQs. Here's some
3:12"Oh, here's the Framer FAQs. Here's some uh official URLs where we can have a
3:14uh official URLs where we can have a
3:14uh official URLs where we can have a quick look at all the different Framer
3:15quick look at all the different Framer
3:15quick look at all the different Framer things." Um and you can you can
3:17things." Um and you can you can
3:17things." Um and you can you can basically set these things up. This is
3:18basically set these things up. This is
3:18basically set these things up. This is inside of the actual agent skills. So,
3:21inside of the actual agent skills. So,
3:21inside of the actual agent skills. So, we have these agents. And we're a little
3:23we have these agents. And we're a little
3:23we have these agents. And we're a little bit we're sloppy on the naming
3:25bit we're sloppy on the naming
3:25bit we're sloppy on the naming convention right now. It's going to get
3:26convention right now. It's going to get
3:26convention right now. It's going to get better and we'll release some guides as
3:28better and we'll release some guides as
3:28better and we'll release some guides as to how to do this. Um but essentially
3:30to how to do this. Um but essentially
3:30to how to do this. Um but essentially right now we have like negative
3:32right now we have like negative
3:32right now we have like negative features. So, no self-hosting. So,
3:34features. So, no self-hosting. So,
3:34features. So, no self-hosting. So, Framer sites are locked to Framer
3:35Framer sites are locked to Framer
3:35Framer sites are locked to Framer hosting. Uh here's the sources where
3:37hosting. Uh here's the sources where
3:37hosting. Uh here's the sources where this is coming from because what you
3:39this is coming from because what you
3:39this is coming from because what you don't want to do uh is put all this
3:41don't want to do uh is put all this
3:41don't want to do uh is put all this information in there that's incorrect
3:43information in there that's incorrect
3:43information in there that's incorrect without the sources, without it being
3:45without the sources, without it being
3:45without the sources, without it being generated. Uh if you have all these
3:47generated. Uh if you have all these
3:47generated. Uh if you have all these sources in there and you're generating
3:49sources in there and you're generating
3:49sources in there and you're generating this stuff, I actually personally think
3:51this stuff, I actually personally think
3:51this stuff, I actually personally think it's better than a human researched
3:54it's better than a human researched
3:54it's better than a human researched system because it knows exactly what
3:56system because it knows exactly what
3:56system because it knows exactly what it's looking for. It knows the fours and
3:57it's looking for. It knows the fours and
3:57it's looking for. It knows the fours and against.
3:59against.
3:59against. Imagine what you can do with this with
4:01Imagine what you can do with this with
4:01Imagine what you can do with this with comparison pages, you know, from one
4:02comparison pages, you know, from one
4:02comparison pages, you know, from one software company to another. Everybody
4:04software company to another. Everybody
4:04software company to another. Everybody does it. Don't get me wrong. It's not
4:06does it. Don't get me wrong. It's not
4:06does it. Don't get me wrong. It's not the most ethically like
4:09the most ethically like
4:09the most ethically like outstanding, really, really nice way of
4:11outstanding, really, really nice way of
4:11outstanding, really, really nice way of doing business, but it is something that
4:13doing business, but it is something that
4:13doing business, but it is something that pretty much everybody does comparison
4:14pretty much everybody does comparison
4:14pretty much everybody does comparison pages.
4:15pages.
4:16pages. Um but yeah, you can see this is a this
4:18Um but yeah, you can see this is a this
4:18Um but yeah, you can see this is a this is an example finding and then you can
4:19is an example finding and then you can
4:19is an example finding and then you can create skills based on these. Here's
4:21create skills based on these. Here's
4:21create skills based on these. Here's like the guardrails. Um so skeptical
4:23like the guardrails. Um so skeptical
4:23like the guardrails. Um so skeptical voices, all kinds of different things.
4:26voices, all kinds of different things.
4:26voices, all kinds of different things. You might want to put, you know, like
4:27You might want to put, you know, like
4:27You might want to put, you know, like this is
4:28this is
4:28this is uh the way that we talk. This is our
4:30uh the way that we talk. This is our
4:30uh the way that we talk. This is our messaging. There's going to be something
4:31messaging. There's going to be something
4:32messaging. There's going to be something if you go dig through this. I bet I've
4:33if you go dig through this. I bet I've
4:33if you go dig through this. I bet I've I've showed something here where I
4:35I've showed something here where I
4:35I've showed something here where I shouldn't have showed it, but you you
4:36shouldn't have showed it, but you you
4:36shouldn't have showed it, but you you get the idea, right? Um
4:38get the idea, right? Um
4:38get the idea, right? Um and so what's interesting about this is
4:41and so what's interesting about this is
4:41and so what's interesting about this is once you've got all this together, you
4:43once you've got all this together, you
4:43once you've got all this together, you can create a loop that basically keeps
4:45can create a loop that basically keeps
4:45can create a loop that basically keeps going and keeps going. You can set it on
4:47going and keeps going. You can set it on
4:47going and keeps going. You can set it on cron jobs. Uh we set a whole bunch on
4:49cron jobs. Uh we set a whole bunch on
4:49cron jobs. Uh we set a whole bunch on cron jobs inside of here. Let me just
4:50cron jobs inside of here. Let me just
4:50cron jobs inside of here. Let me just move this aside and find it. Uh where
4:53move this aside and find it. Uh where
4:53move this aside and find it. Uh where are we? There is a cron in here
4:55are we? There is a cron in here
4:55are we? There is a cron in here somewhere. Schedules. So you can see CMS
4:58somewhere. Schedules. So you can see CMS
4:58somewhere. Schedules. So you can see CMS freshness, for example, is on the cron
4:59freshness, for example, is on the cron
4:59freshness, for example, is on the cron job.
5:00job.
5:00job. Um and I can never read these things,
5:03Um and I can never read these things,
5:03Um and I can never read these things, but this is this is on the I think once
5:04but this is this is on the I think once
5:04but this is this is on the I think once a week and I think it's on a Thursday
5:06a week and I think it's on a Thursday
5:06a week and I think it's on a Thursday from memory. Um but essentially every
5:09from memory. Um but essentially every
5:09from memory. Um but essentially every time it goes and runs through that cron
5:10time it goes and runs through that cron
5:10time it goes and runs through that cron job, you're basically updating across
5:13job, you're basically updating across
5:13job, you're basically updating across all the different CMSs that information.
5:16all the different CMSs that information.
5:16all the different CMSs that information. So you've actually basically created a
5:18So you've actually basically created a
5:18So you've actually basically created a self-updating bot. And guess what the
5:20self-updating bot. And guess what the
5:20self-updating bot. And guess what the bot does after it? It tells you this
5:22bot does after it? It tells you this
5:22bot does after it? It tells you this information. And so once you've built
5:25information. And so once you've built
5:25information. And so once you've built one of these things end to end, it then
5:27one of these things end to end, it then
5:27one of these things end to end, it then becomes very simple to create the next
5:29becomes very simple to create the next
5:29becomes very simple to create the next one and the next one and the next one.
5:32one and the next one and the next one.
5:32one and the next one and the next one. And you can either do it in two
5:33And you can either do it in two
5:33And you can either do it in two different ways. You can either make them
5:34different ways. You can either make them
5:34different ways. You can either make them in isolation, so you can have all these
5:36in isolation, so you can have all these
5:36in isolation, so you can have all these different agents that don't really talk
5:37different agents that don't really talk
5:37different agents that don't really talk to each other, or you can do like we've
5:39to each other, or you can do like we've
5:39to each other, or you can do like we've done here where we have, you know, like
5:41done here where we have, you know, like
5:41done here where we have, you know, like competitor intelligence, we have daily
5:43competitor intelligence, we have daily
5:43competitor intelligence, we have daily AEO, we have a Reddit scout, uh we have
5:46AEO, we have a Reddit scout, uh we have
5:46AEO, we have a Reddit scout, uh we have like root strategies, uh CMS sweeps, all
5:49like root strategies, uh CMS sweeps, all
5:49like root strategies, uh CMS sweeps, all kind of stuff. Like there's a lot of
5:50kind of stuff. Like there's a lot of
5:51kind of stuff. Like there's a lot of information that's in here. Um and
5:53information that's in here. Um and
5:53information that's in here. Um and again, a lot of different skills.
5:55again, a lot of different skills.
5:55again, a lot of different skills. So, I'll show you just a very quick
5:57So, I'll show you just a very quick
5:57So, I'll show you just a very quick example inside of our Slack. I'm trying
5:58example inside of our Slack. I'm trying
5:59example inside of our Slack. I'm trying my hardest to not accidentally like docs
6:01my hardest to not accidentally like docs
6:01my hardest to not accidentally like docs anything or put anything out there, but
6:03anything or put anything out there, but
6:03anything or put anything out there, but if you look inside of Slack here, this
6:05if you look inside of Slack here, this
6:05if you look inside of Slack here, this is a great example of it. So,
6:07is a great example of it. So,
6:07is a great example of it. So, inside of here, we actually have a
6:09inside of here, we actually have a
6:09inside of here, we actually have a website report bot. And this website
6:11website report bot. And this website
6:11website report bot. And this website report bot takes information from Vercel
6:14report bot takes information from Vercel
6:14report bot takes information from Vercel observability and a whole bunch of other
6:16observability and a whole bunch of other
6:16observability and a whole bunch of other different touch points and basically
6:18different touch points and basically
6:18different touch points and basically tells us what's going off. And because
6:20tells us what's going off. And because
6:20tells us what's going off. And because we've got all this information ahead of
6:21we've got all this information ahead of
6:22we've got all this information ahead of us, we can actually make educated
6:24us, we can actually make educated
6:24us, we can actually make educated decisions as to how we progress the
6:26decisions as to how we progress the
6:26decisions as to how we progress the website. So, this is a really cool
6:29website. So, this is a really cool
6:29website. So, this is a really cool feedback loop and it's been pretty much
6:31feedback loop and it's been pretty much
6:31feedback loop and it's been pretty much vital for our growth as a business,
6:34vital for our growth as a business,
6:34vital for our growth as a business, where you can continuously spin these
6:36where you can continuously spin these
6:36where you can continuously spin these things up, continuously rely on these.
6:39things up, continuously rely on these.
6:39things up, continuously rely on these. And the best bit of it all is I put like
6:41And the best bit of it all is I put like
6:41And the best bit of it all is I put like $100 on our AI
6:44$100 on our AI
6:44$100 on our AI whatever it is, the AI gateway.
6:47whatever it is, the AI gateway.
6:47whatever it is, the AI gateway. And I haven't spent it yet. And we've
6:49And I haven't spent it yet. And we've
6:49And I haven't spent it yet. And we've got like five different
6:51got like five different
6:51got like five different agents that are constantly running. And
6:53agents that are constantly running. And
6:54agents that are constantly running. And we've been doing this since pretty much
6:55we've been doing this since pretty much
6:55we've been doing this since pretty much day one of Eve dropping. Um not Eve
6:58day one of Eve dropping. Um not Eve
6:58day one of Eve dropping. Um not Eve dropping, but like Eve the tool
7:00dropping, but like Eve the tool
7:00dropping, but like Eve the tool dropping.
7:01dropping.
7:01dropping. Um and so it's yeah, it's it's pretty
7:03Um and so it's yeah, it's it's pretty
7:03Um and so it's yeah, it's it's pretty pretty wild. It's pretty cool.
7:05pretty wild. It's pretty cool.
7:05pretty wild. It's pretty cool. And I would definitely recommend that
7:06And I would definitely recommend that
7:07And I would definitely recommend that this is going to be the way that people
7:08this is going to be the way that people
7:08this is going to be the way that people use agents more with being able to take
7:11use agents more with being able to take
7:11use agents more with being able to take all these different disparate systems,
7:14all these different disparate systems,
7:14all these different disparate systems, pile them into a single Slack channel
7:16pile them into a single Slack channel
7:16pile them into a single Slack channel and not making it overwhelming. Um there
7:19and not making it overwhelming. Um there
7:19and not making it overwhelming. Um there is a different way of doing this
7:20is a different way of doing this
7:20is a different way of doing this completely separately as well, which is
7:21completely separately as well, which is
7:21completely separately as well, which is kind of interesting, which is you can
7:23kind of interesting, which is you can
7:23kind of interesting, which is you can use things like Claude's schedule, but
7:25use things like Claude's schedule, but
7:25use things like Claude's schedule, but the problem with this is is Claude
7:26the problem with this is is Claude
7:26the problem with this is is Claude scheduled and I can talk about this in
7:28scheduled and I can talk about this in
7:28scheduled and I can talk about this in another video if anybody's interested,
7:30another video if anybody's interested,
7:30another video if anybody's interested, but Claude scheduled is a
7:32but Claude scheduled is a
7:32but Claude scheduled is a non-deterministic way of doing this and
7:34non-deterministic way of doing this and
7:34non-deterministic way of doing this and it's pretty difficult to get it to be
7:36it's pretty difficult to get it to be
7:36it's pretty difficult to get it to be deterministic. So, it means that you get
7:38deterministic. So, it means that you get
7:38deterministic. So, it means that you get kind of like instead of rigid things
7:41kind of like instead of rigid things
7:41kind of like instead of rigid things where it tells you and and being able to
7:43where it tells you and and being able to
7:43where it tells you and and being able to kind of piecemeal get better and better.
7:46kind of piecemeal get better and better.
7:46kind of piecemeal get better and better. It's kind of like a scatter shot where
7:48It's kind of like a scatter shot where
7:48It's kind of like a scatter shot where it might answer this way one week and it
7:50it might answer this way one week and it
7:50it might answer this way one week and it might answer this way another week. And
7:53might answer this way another week. And
7:53might answer this way another week. And the problem is it's very difficult then
7:55the problem is it's very difficult then
7:55the problem is it's very difficult then to, you know, take that information week
7:57to, you know, take that information week
7:57to, you know, take that information week by week and compare it to one another if
8:00by week and compare it to one another if
8:00by week and compare it to one another if it doesn't match the same data
8:01it doesn't match the same data
8:02it doesn't match the same data structure. So,
8:03structure. So,
8:03structure. So, I'm I'm pretty heavily I'm I I hate this
8:06I'm I'm pretty heavily I'm I I hate this
8:06I'm I'm pretty heavily I'm I I hate this expression, but I'm pretty bullish on
8:09expression, but I'm pretty bullish on
8:09expression, but I'm pretty bullish on Eve being like a long-term
8:11Eve being like a long-term
8:11Eve being like a long-term uh tool that we use internally to take
8:13uh tool that we use internally to take
8:14uh tool that we use internally to take all these different systems, mesh them
8:15all these different systems, mesh them
8:15all these different systems, mesh them together, and just tell me what the
8:17together, and just tell me what the
8:17together, and just tell me what the hell's going off on the website rather
8:19hell's going off on the website rather
8:19hell's going off on the website rather than being completely clued out and not
8:20than being completely clued out and not
8:20than being completely clued out and not understanding how, you know, like Google
8:22understanding how, you know, like Google
8:22understanding how, you know, like Google Analytics works, Google Search Console
8:24Analytics works, Google Search Console
8:24Analytics works, Google Search Console works, uh Posthog, Ahrefs, whatever
8:27works, uh Posthog, Ahrefs, whatever
8:27works, uh Posthog, Ahrefs, whatever different disparate systems. You get the
8:29different disparate systems. You get the
8:29different disparate systems. You get the idea. Anyway, if you want more info
8:31idea. Anyway, if you want more info
8:31idea. Anyway, if you want more info about this,
8:32about this,
8:32about this, drop us a comment down below and we'll
8:34drop us a comment down below and we'll
8:34drop us a comment down below and we'll we'll get back to you. Thanks for
8:36we'll get back to you. Thanks for
8:36we'll get back to you. Thanks for watching. Have a great day. Take care.
8:38watching. Have a great day. Take care.
8:38watching. Have a great day. Take care. [music]
8:38[music]
8:38[music] Bye-bye.
Questions answered
Editorial answers from the original Roboto page.
What is Eve and what can it be used for?
Eve is a platform for building AI agents that can operate across different systems. You can use it to create chatbots in Slack, automate website changes, build comparison pages, or integrate with tools like Linear and GitHub. The key advantage is that Eve lets you connect these agents directly to whatever tools you're already using through Vercel Connect.
Read the original ↗How does Eve help with keeping comparison pages up to date?
We use Eve with a YAML file-based system that stores all our comparison page information in one place. This makes it versionable and allows us to create pull requests to update content. Eve can automatically run these updates on a schedule using cron jobs, so our comparison pages stay fresh without manual intervention.
Read the original ↗Why is using deterministic systems better for AI agents?
Deterministic systems are much easier and cheaper for AI to work with because they follow predictable patterns and rules. When you set up your agent skills with clear guardrails and structured data like we do, the AI can reliably execute the same tasks repeatedly, which significantly reduces your token costs.
Read the original ↗How much does it cost to run multiple agents with Eve?
We've set up five different agents that are constantly running across our systems, and we put a hundred dollars on our AI Gateway at the start. We haven't spent it yet, so the cost is remarkably low for what you get in terms of continuous automation and intelligence gathering.
Read the original ↗How can Eve integrate with Slack?
Eve connects directly to Slack through Vercel Connect, allowing you to set up agents that send information and reports into your Slack channels. For example, we have a website report bot that pulls data from Vercel observability and other sources, then delivers insights right to Slack so your team can make decisions based on real-time information.
Read the original ↗What makes Eve better than non-deterministic approaches like Claude Schedule?
Non-deterministic systems like Claude Schedule can give different answers each week depending on how the AI interprets things, which makes it hard to track changes over time or compare data. Eve's deterministic approach means your agents return consistent data structures, so you can reliably compare week-to-week results and actually see what's improving or degrading on your website.
Read the original ↗Can I run multiple agents together or do they need to be separate?
You can do both. You can create isolated agents that work independently, or you can build connected systems where multiple agents feed into each other. We've built something more integrated with agents like competitor intelligence, SEO tracking, Reddit scouting, and CMS updates all working together and reporting into a single Slack channel.
Read the original ↗Scope and context
- Video authors are credited by the original Roboto page; individual transcript speakers remain unassigned.