[{"data":1,"prerenderedAt":67},["ShallowReactive",2],{"leap-week-personalization-at-scale":3,"leap-week-personalization-at-scale-next":53},{"id":4,"slug":5,"vimeo_id":6,"description":7,"tile":8,"length":9,"resources":10,"people":10,"episode_number":11,"published":12,"title":13,"video_transcript_html":14,"video_transcript_text":15,"content":10,"status":16,"episode_people":17,"recommendations":36,"season":37,"seo":52},"1310befc-e361-4e19-848f-d685c19dddef","personalization-at-scale","1176544348","Learn how to use enrichment data to build hyper-personalized landing pages at scale. Featuring Clay.\n\n","f736efe0-fb5d-4453-8cde-408ba3643b8c",27,null,6,"2026-03-27","Personalization at Scale","\u003Cp>Speaker 0: Hey. What's up, everybody? Matt here from the direct us marketing team. I am super excited today to be showing off, what I think is a really cool session personally. Kinda set the scene a little bit.\u003C\u002Fp>\u003Cp>Obviously, I said marketing, and that's probably terrifying for a lot of people. But, I I like the I I truly believe the gap between, like, technical and nontechnical is getting, you know, smaller. So Mhmm. That's why I'm super excited to have Clay and my friend Mohawk Desai here. Mohawk, how are you, man?\u003C\u002Fp>\u003Cp>Speaker 1: Hey. How's it going, Matt? Thanks for bringing me on. Appreciate it.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Yeah. Absolutely. So most of our audience is probably not familiar with Clay. Yeah.\u003C\u002Fp>\u003Cp>I will say, like, I I think Clay was kind of the first step for me to get to become more technical, with a lot of enrichment and sort of stuff, but like that. But maybe you could give kind of a a brief Yeah. Overview of what Clay is for the folks.\u003C\u002Fp>\u003Cp>Speaker 1: Yeah. Absolutely. So at a high level, Clay is a a GTM orchestration and data enrichment tool. And I think a good analogy is I would view Clay as almost a development environment, except not for technical folks, not for coders, but for GTM nerds, like me and like Matt. So, yeah, Clay is if you look at Clay, and we'll jump into the product in a second, but it looks like Excel.\u003C\u002Fp>\u003Cp>There's a lot of AI formulas. There's a lot of, sort sort of technical concepts like webhooks and HTTP APIs. But it is really built for you know, right now, one of our biggest personas is, like, rev ops people at companies, but that's kind of expanding now into marketing folks, into mark ops, and to other personas down the road as well, including, like, sales. But it's really for, like, systems thinkers who wanna build their sort of GTM engine. Clay is kind of the perfect middle ground for all of that, where you can have your entire tech stack flow into Clay.\u003C\u002Fp>\u003Cp>We integrate with 90 plus providers, for data, but as well as sort of GTM execution tools like Gong. And so Clay is kind of like the glue between all your different tools, as well as a place where you could secure data, the things like emails and phone numbers. So, really, it's sort of a one stop shop for GTM orchestration. It's probably the best way I can describe Clay.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Wow. That's a you're reading off like a prompter there. That was awesome. No.\u003C\u002Fp>\u003Cp>The way that you yeah. The way you said, like, it looks like a spreadsheet when you first get into it, that was, like, my first reaction when I first started testing it. You know, I was like, what? This just looks like another spreadsheet. And then when you start looking at the integrations and the possibilities with, like, enrichment waterfalls, that's what, like, had my holy But speaking of integrations, you know, Directus is, you know, we're we're we've always been kind of a content management system for a lot of folks.\u003C\u002Fp>\u003Cp>So all of the pages for your website, it could be the back end for applications. There's a lot of different use cases. And as we've been getting Clay and Directus closer together, our team has found a lot of really cool use cases that you can really dive into, with that enrichment data. One of which I'm really excited to talk about today. And I feel like was this, like, Clay's, like, bread and butter early on, this sort of, like, personalization type play?\u003C\u002Fp>\u003Cp>Speaker 1: I would say not early on. I think, actually, early on in the moat was this the data enrichment piece. On the back end to find emails and phone numbers, Clay takes a waterfall approach where we don't just check one database, but we'll cycle through 90, again, 90 plus providers to find an email or phone number. And so that was the note early on was a higher coverage rate and higher accuracy for data versus under other individual vendors. But, Matt, to your point, I think in the past year or so, a lot of smart rev ops people and marketing people are realizing that Clay's a really good place to do ABM at scale, so account based marketing at scale.\u003C\u002Fp>\u003Cp>And one of them is, yes, starting to create personalized landing pages at scale for your target accounts. But this can also leak into other cool use cases like you can create Google Slides at scale that do the same thing. We have a direct integration of Google Slides, and and so on and so forth. And now in the in the age of AI with all the tools available, we're trying to be a little bit more open to to tools that will allow you to do this. But Directus is, like, a perfect, perfect example of a great integration where we can, really customize our outreach and get those target accounts in our book.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Yeah. And when you say ABM, I'm sure the the audience is like, another acronym, like marketing and some of the acronyms. But, for folks that don't know, so ABM is is kind of an approach to marketing and sales that, in my opinion, is the way to do it because it's way less intrusive than just, like, shotgun blasting a bunch of people that have no idea who you are. There's a lot of research and, personalization that goes into identifying, you know, 50 accounts, a 100 accounts that are assigned specifically to, like, your go to market team or your reps.\u003C\u002Fp>\u003Cp>So they get to build relationships and really understand, like, the pains and and things within those those companies. And that kinda sets the scene for today, which is, we're gonna do a workflow from beginning to end and and show you how we built it, basically, where it's not like those shitty emails you'll get that's like, hey. Like, I saw you play volleyball, and now let's get on a fifteen minute call. Like, it that's not personalization. Like, personalization is looking at, like, who a company is hiring, you know, what teams are growing, what is probably because of that, some of the struggles they have right now that your product or solution could maybe solve for, and then building out pages and stories around that.\u003C\u002Fp>\u003Cp>But that's hard to do without automation, without things like Clay for the data enrichment. So that's what I'm excited to talk about today.\u003C\u002Fp>\u003Cp>Speaker 1: Amazing. Let's do it.\u003C\u002Fp>\u003Cp>Speaker 0: Awesome. Alright. So let's hop in. I hope everything is gonna break. I know as soon as I start sharing stuff.\u003C\u002Fp>\u003Cp>It always I don't know why. It always happens. But, let's start with, like, kind of what we're gonna have at the end. That that's kind of our our end goal. We'll start there and work backwards.\u003C\u002Fp>\u003Cp>So, this is just a quick landing page that I spun up, like, ten minutes before this, with clogged code. UIs, you know, you you can spin up a UI really fast now. But the goal is to have this fake AI agency that we just made up called Cottontail, then AI operations team. So they'll handle a lot of the, the manual processes and things like that. So we're gonna have these customized landing pages where it's like, this is the typical personalization you see.\u003C\u002Fp>\u003Cp>Like, all your company name, your, you know, subheader. But what I'm really excited about is focusing in on this section, which is, like, we're gonna use Clay to basically pull out, you know, who are the the number the highest number of jobs, what teams are really growing, and run a couple of, you know, flows on, what does that mean? Like, what are those struggles, the problems they probably have because of that? And then how can our solution fix that? So, basically, a mirror image of what I just said, but this is what it looks like in practice.\u003C\u002Fp>\u003Cp>Any thoughts on, like you know, obviously, working at Clay, like Yeah. Clay, in my opinion, is one of the best marketing. They're I they're like the shining pinnacle right now, everything that y'all do. Mhmm. And I wanna say ABM is a big part of that.\u003C\u002Fp>\u003Cp>Right?\u003C\u002Fp>\u003Cp>Speaker 1: Or it could be cool\u003C\u002Fp>\u003Cp>Speaker 0: at all.\u003C\u002Fp>\u003Cp>Speaker 1: No. Absolutely. As you mentioned, ABM is becoming even a a larger and larger segment of a use case at Clay. I think typically, customers will come in and start with basic enrichment, basic, automated outbound or automated inbound, but those really savvy, you know, companies that know they have sort of a target list of a hundred, two hundred, you know, large companies or or target accounts. ABM is really, really effective at breaking in when everything else doesn't work.\u003C\u002Fp>\u003Cp>And, our CEO, Kareem, has a very cool, analogy or a sort of phrase that encapsulates what ABM should look like, and that's surround sound. You're surround sound of your customers, in in a way that feels so personalized that they they feel obligated to sort of respond or at least curious enough to alright. What do these guys have to say to us? So ABM is like there's a lot that goes into it. One of them is this use case we're gonna jump into, which are these, like, landing pages.\u003C\u002Fp>\u003Cp>Matt, you're correct. This is becoming a very popular thing amongst advanced GTM teams. But outside of just custom landing pages, there's a lot of other things that kind of fold up into ABM. One of them is signals. So taking a signal based approach and having your reps work off intent based data rather than static third party data.\u003C\u002Fp>\u003Cp>So that's one piece. And then the third piece could also be ads. Clay also just launched a new ads tool, but in general, ads is just another way to sort of, again, surround sound, your your personas at these companies. So now imagine a head of marketing at one of your target accounts is getting brilliant, custom landing pages, great emails from your sales teams, are are getting, you know, timely outreach based on some funding events or relevant events to your company, and then you're also seeing ads on LinkedIn and Meta, it's gonna be hard to not be curious about what you do. So that's ABM in a nutshell.\u003C\u002Fp>\u003Cp>And, Matt, I think you're right that that this is picking up a lot in the market.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Yeah. That's that's awesome. And when you start folding in, like, billboards and buses with your Yes. Exactly.\u003C\u002Fp>\u003Cp>It's inescapable, man.\u003C\u002Fp>\u003Cp>Speaker 1: Yes. Inescapable. Exactly. Exactly.\u003C\u002Fp>\u003Cp>Speaker 0: But in a, like, in a way, I think it's it's better than just kind of blindfold just going out there, like, blasting you know, pulling a list of a 100,000 contacts and, like, spray and pray method.\u003C\u002Fp>\u003Cp>Speaker 1: Spray. Exactly.\u003C\u002Fp>\u003Cp>Speaker 0: So, so yeah. So that's what we're gonna do today is really focus on on building these pages out, which I think is really the foundation of it. This is not only something for the customers, but also, you know, for your team members as they start to do the research having, like it's basically like a slide deck in in a lot of ways. So just to kind of show you how I've set up direct us in the back end, so we've got, Cottontail here. I've created two collections.\u003C\u002Fp>\u003Cp>So we've got a landing pages collection. This is where all of the stuff we're gonna do in clay is gonna push, right here. So looking at the actual data model schema for this, it's it's basically everything I showed on that that front end page. So Mhmm. Company name, we're gonna have slug, so this will, like, auto generate the pages when they they come in.\u003C\u002Fp>\u003Cp>Everything is powered by by Railway, which is really the one of the easiest ways to get started with Directus and and build front ends on. We got industry. We got a little testimonial thing. We'll see if we have enough time. I I definitely wanna show how the mini to one, like, data schema stuff works.\u003C\u002Fp>\u003Cp>And then we'll have the personalization content kind of backfilled here. So, like, as developers or technical people are watching this, like, this is how you can knock something out. Like, if you do this for your marketing team, it would be, like, you're you're the hero for for months just giving them something like this. But, also, if you're building applications or products, like, this is a great way to get it into the market and start, you know, marketing the stuff that you're building. But yeah.\u003C\u002Fp>\u003Cp>So that's all built out. Just to show you what it would look like if you were to manually fill this out, yeah, just like a a typical form here. This is just how I have it laid out, but you can do it a lot of different ways. But, but, yeah, I that's really all I had for the director side. Like, as a back end, it's not sexy.\u003C\u002Fp>\u003Cp>It's just like we need to push all this data here, but I'm really excited to show what we've got for Clay. So, Mohawk, I'll actually turn it over to you.\u003C\u002Fp>\u003Cp>Speaker 1: Amazing. Thank you, Matt. Great intro. Great context. So now what I'm gonna do is share my screen, and go into Clay.\u003C\u002Fp>\u003Cp>So for folks that have never seen Clay, welcome to Clay. This is the spreadsheet looking tool that you keep hearing about, and so this is what it looks like. So as a starting point for this example, we've uploaded, as our first step, a list of company domains, just their websites. In Clay, though, you can import these company domains from a bunch of other sources. So we have our own company search where you can filter, you know, by, industry, location, revenue count, etcetera.\u003C\u002Fp>\u003Cp>So if you don't already have a list of target accounts in your CRM or in a CSV, you can use Clay to find those target accounts based on your filters. And, as I mentioned, you can upload a CSV as well or connect your CRM and have a live sync between Salesforce or HubSpot, and and Clay. But, again, zooming back out, we are starting with a list of company domains that we have randomly chosen and put into this table. That's the only input we're putting in. That's the I think the key point of all this is that all these other columns you see that have been set up, you set them up once, and anytime you add a new input, it'll automatically run for, for that new input.\u003C\u002Fp>\u003Cp>So, actually, Matt, I I I'm gonna go ahead and sort of enter a new domain here. And, this is my favorite company in the world. So Apple, we're gonna put that a bit second to play. And so I've put this in, and as you can see, it's already autofilled. And I'll kind of now go column by column and break down what we did here to get to that landing page endpoint that Matt showed.\u003C\u002Fp>\u003Cp>So the the first thing is we ran this action called enrich company, and that essentially, in in a JSON array, will return a bunch of data about the company that could be useful. So things like their employee count, their revenue range, their, again, their their, country of origin, things like that. And so the the relevant data points we wanted to pull out for that landing page for each of these companies are really just the company name, the slug, the employee count, and the industry. What we also got, which is pretty cool, and this is kind of taking customization to that, like, final level, is we have URLs to the company's logo as well as a screenshot of their website. So now we've imported their their brand, what they look like, what they feel like, and that can be used if you'd like in your custom landing page, that you're sending to these target accounts.\u003C\u002Fp>\u003Cp>Then from there, we look into jobs. Now jobs in general, I think, are very underrated data point that GTM team should be indexing on because jobs tell you a lot about the company. Tells you what their tech stack is for specific roles. Right? Like, what Mhmm.\u003C\u002Fp>\u003Cp>In that little section that says you should have experience with these technologies, it tells you what they're using. It tells you about their sort of their mission, what they're up to, kind of everything that a candidate would need to know happens to be a lot of the information that we wanna know as sellers and marketers. So, we went ahead and ran an action called find active job openings at this company. And, all all we really needed here was the domain as the input. Again, we're working off of this one input.\u003C\u002Fp>\u003Cp>It's really all we need. And so once you put your domain into this action, that we, you know, we set sort of our filters and we run it, it'll return how many jobs are sort of posted for this as well as the link to each job, which is wonderful because we can then run an AI column on this, active job opening column to kind of analyze those job postings and extract the key insights that we want, like I mentioned. And so we're able to find, which teams are growing fastest, which functions appear to be manual or understaffed. And imagine replacing all of this with the pain points you wanna extract about these companies from their job postings. And so this is this is where we start going into, like, combining the power of AI with the power of of Clay's customization, and then finally that that final layer of direct is coming in as that that content layer.\u003C\u002Fp>\u003Cp>So we've extracted into columns, sort of these different pain points slash operational insights. I shouldn't say pain point, actually. These are specifically how is the team operating, and they're stored in columns, which we'll use to upload that into those, different fields that Matt showed at the beginning. So great. We have our operational insight columns built out.\u003C\u002Fp>\u003Cp>We also use AI to develop, like, personalized headers and sub headers for that landing page. So, essentially, all the inputs we're putting into Directus are all developed in Clay call my call. And so we got personalized sub headers, which are really cool. And then we finally got to those pain points, which I was a little too excited to talk about. But, if we go here to generate, you'll kinda see the prompt, but we're giving it business context.\u003C\u002Fp>\u003Cp>We're giving it, what we do as an agency, as a fake agency. And then it we tell the AI simply what we would tell maybe a new intern or a new hire company, and and we kind of talk to the AI exactly like that, and we get great results. So we're saying, right, three pain points on their operations or go to market team using the inputs from these outputs that we found before. And, for each one kind of return just like a a quick one liner that we can upload to the website. And so this is an example of the final final result here.\u003C\u002Fp>\u003Cp>So that's all the customization we've done in Clay, and the final step to return this data back to Directus is using HTTP API call. Now when I first joined Clay, I'll be honest, I had no idea what this was. And it sounds complicated and scary, but it is a very we try to make it as easy as possible in Clay, and we literally tell you what to copy and paste in here. But this is essentially just a way to share data between the Clay table and Directus. And so in this case, there's a few methods, to get it's it's pretty self explanatory, actually.\u003C\u002Fp>\u003Cp>If you're getting data, that means you're returning data into this table. But in this case, we're posting data, sort of pushing data out. So you selected that as our method. Matt provided me an endpoint from Directus that we can then paste into here. And then, in our body, this was actually, Matt, you wanna Matt actually set this column up.\u003C\u002Fp>\u003Cp>Matt, how did you generate this this sort of body?\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. I used AI.\u003C\u002Fp>\u003Cp>Speaker 1: So, like I was I was about to say that. Yeah.\u003C\u002Fp>\u003Cp>Speaker 0: Perfect. Yeah. No. Yeah. I mean, one of the that this was actually my first time setting this up, and, our, integrations engineer, Lynn Lindsay, has actually built a Directus specific Directus CMS HTTP API.\u003C\u002Fp>\u003Cp>So if you go to, like, set up an enrichment column, just search Directus and it'll pop up, and there's some instructions in there. But I we also have some docs, which I'll show once we hop over from here, of where you can find those. We also talk about how to get it set, set up on our side as well.\u003C\u002Fp>\u003Cp>Speaker 1: Yep. Yeah.\u003C\u002Fp>\u003Cp>Speaker 0: So the item from get create update. Yeah.\u003C\u002Fp>\u003Cp>Speaker 1: Yeah. Exactly. So it's kind of alluding to all the hard work Matt did to set up this column. He's made it super simple and saved this as a public template that you'll be able to access. If you search up direct as HTTP in the clay search bar, you'll see this exact enrichment here, which is the get item from CMS collection.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Yeah. Great. And, basically, to to to for the, for the JSON, for this, you know, be able to pull this into Directus, what I did when I say I used AI, like, I literally just took a screenshot of the data model from the Directus side that I showed earlier and then just dropped it in a cloud and was like, can you create, like, a JSON placeholder for Glay? And then Beautiful.\u003C\u002Fp>\u003Cp>My my favorite part actually is, like, mapping the actual columns to like, for some reason, it's just like it's like Legos. Like, I just love Yeah. Yeah. Slash then just, like, go through and pick it. Yeah.\u003C\u002Fp>\u003Cp>It's very satisfying.\u003C\u002Fp>\u003Cp>Speaker 1: Yeah. That's by the way, Legos for adults or Minecraft for adults is the analogy I use when I think about clay. A big reason why I joined the company. I get to work with Legos every day now. Yeah.\u003C\u002Fp>\u003Cp>Cool. So that that's that's the last step of the Clay workflow. All this data now, because of Matt's hard work with this HTTP API call, flows back in direct us. I will now pass it off back to Matt to show the final product.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Did, did you run those columns again? All five.\u003C\u002Fp>\u003Cp>Speaker 1: Yeah. All all six, actually. So Apple should have run as well. Yeah.\u003C\u002Fp>\u003Cp>Speaker 0: Let me see.\u003C\u002Fp>\u003Cp>Speaker 1: So I'll you want me to rerun them real quick?\u003C\u002Fp>\u003Cp>Speaker 0: Toss it over. Toss the screen over. Okay. I'll just share my screen and see, and we'll just, like Cool.\u003C\u002Fp>\u003Cp>Speaker 1: Cut out as necessary. Yeah.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. We'll, like, we'll do it live. Alright. So we have our direct distance here. If I refresh, everything went according to plan.\u003C\u002Fp>\u003Cp>So it looks like Apple got pushed over. So if you, on your side, run that Yeah. HTTP\u003C\u002Fp>\u003Cp>Speaker 1: Call call. Gotcha.\u003C\u002Fp>\u003Cp>Speaker 0: Again. Yes. So those should come in one by one. And while you're doing that so like I said, all of the data is automatically pushed over. I I didn't have to lift a finger for this.\u003C\u002Fp>\u003Cp>Where it gets interesting now is, like, we have the logo URL, so we can build, you know, like a we have, flows and directives, which I haven't really talked about much, but you can set up, automations from the data inside here. So we can run a flow that's like, hey. Like, take that URL, go and fetch the the logo, and then you'll have, like, you know, all of that to personalize on as well. So, it looks like everything is in here. Yeah.\u003C\u002Fp>\u003Cp>So it looks like all of our other ones got pushed as well. Cool. I mentioned I'll just show you really quick. So if you go to our docs and with that HTTP API that we set up. So we have in here, integrations page.\u003C\u002Fp>\u003Cp>So you can just come in and and search search search Clay, and you should be able to find, yeah, quick and easy setup, how to use, like, the templates and stuff here. And we tell you, everything pretty well. Shout out to Lindsay for for her work on all this. But cool. Alright.\u003C\u002Fp>\u003Cp>So let's see if it works. That's probably Yeah. Probably the most important part of this. Yeah. Alright.\u003C\u002Fp>\u003Cp>So we've got this. We've got the slug, which should auto apply, to the pages. I am gonna pop in. So I I mentioned this earlier, but we have, like the cool thing about the data in Directus is when you create these collections, you can do, like, many to one so you can associate separate collections. So you have a team that just has access to collections with, like, well based access control, and you can have a team that just has access to landing pages.\u003C\u002Fp>\u003Cp>Speaker 1: But they\u003C\u002Fp>\u003Cp>Speaker 0: can Nice. Their work can, like, counter, like, counteract. So or in a good way, not not in a bad way. But, like, for Apple, right, like, I don't know if we have I guess we can pull a SaaS in. So we would wanna pull in, like, the specific SaaS testimonial here, which you could probably trigger with another flow, but I'm not that advanced.\u003C\u002Fp>\u003Cp>But I also like to do some manual, manual work around this. So let's see. Moment of truth. So coming over to our front end here.\u003C\u002Fp>\u003Cp>Speaker 1: Got\u003C\u002Fp>\u003Cp>Speaker 0: basic page. Type in apple.\u003C\u002Fp>\u003Cp>Speaker 1: There you go.\u003C\u002Fp>\u003Cp>Speaker 0: Apple. Beautiful. Ops team. Wow. So we've got our subheader, All of our specific problems where it looks like they're hiring a lot of engineering, so we can for our products, you know, intake and routing workflows, that support the onboarding, that sort of stuff.\u003C\u002Fp>\u003Cp>Yeah. And then our our SaaS testimonial here. So it's like I mean, if if you take out our talking, what? This is probably, like, a two to five minute workflow if you have, like, all of the basic pieces, and you can scale ABN that fast, which with, like, personalization. That's actually good personalization.\u003C\u002Fp>\u003Cp>I like that that's what has always, like, impressed me with this is just how granular you can get and refine and, like, now you have this. Keep iterating, testing, see what resonates.\u003C\u002Fp>\u003Cp>Speaker 1: Yes. And I think, Matt, I'll say one thing here. With great power comes great responsibility. I think it's still on the person that's developing the workflow to have good taste. Good taste in what type of personalization do you do versus what you don't.\u003C\u002Fp>\u003Cp>Matt alluded to a great point earlier. Maybe don't mention their kids' Facebook, you know, profile or something random about them that that they're not gonna care about. It sounds contrived. People are tired of getting AI sloppy emails and over personalized messages. So there's a fine line of personalizing tastefully with something like this versus sort of being, having something contrived that's not gonna work.\u003C\u002Fp>\u003Cp>But, nonetheless, the technology is here. It's exciting, and now we're we're really happy to put this in in your hands.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Yeah. And that's like you just said, like, alluding back to when I said, like, when I said I'm in marketing and people probably cringed and run out, like, the garlic and holy water.\u003C\u002Fp>\u003Cp>Speaker 1: Like,\u003C\u002Fp>\u003Cp>Speaker 0: we we try to do it the right way. Like, we're we're trying to be there's, like, good marketing and bad marketing. And I tell, like, our engineering team all the time, like, I'm I'm trying to change the way that people think about it. Because a lot of, like Mhmm. A lot of it has been just, like, those quick growth hack things that I hate.\u003C\u002Fp>\u003Cp>But, when you do things like this and you actually start to build relationships and understand and show that, like, you are doing the work to learn about Mayo Clinic and, you know, some of the pains and struggles they have. Like, that's it's not contact level. You don't have to worry about PII. You're scraping all the available data you can find on the site. So, yeah, I I think this is you know, we keep alluding to it, but the way forward and, it's cool to see that Clay and Directus, like, just as a stack can can do it.\u003C\u002Fp>\u003Cp>Yeah. Obviously, AI helping a lot of things, but, I say it all the time. Everybody's becoming a builder, and you're probably seeing this over at Clay too. Like Yep. Them and moat.\u003C\u002Fp>\u003Cp>Like, this used to be, like, for all the developers in this call, probably a pain in the butt because you had marketing asking, like, can you create this page for me? Can you create this state? Like, how do I get this data here? And like I said, I've opened terminal before I worked here at Directus probably twice in my life, and now using cloud code. Being able to do all this myself and not have to bother anybody, that's the way things are shifting for for us as, you know, go to market team.\u003C\u002Fp>\u003Cp>And I'm I'm really excited to see, you know, where we land with it.\u003C\u002Fp>\u003Cp>Speaker 1: Absolutely.\u003C\u002Fp>\u003Cp>Speaker 0: Well, cool, man. Mohawk, I really appreciate you coming on today, giving us\u003C\u002Fp>\u003Cp>Speaker 1: Likewise.\u003C\u002Fp>\u003Cp>Speaker 0: You know, a quick view of Clay, how everything works. I'm so glad everything worked out. We didn't have any technical issues on either end. But, for everybody watching, like, all of the prompts that, were shared, I'll make sure that it's in, like, a accessible place. So if you wanna just grab those and test it yourself.\u003C\u002Fp>\u003Cp>Anything else from this call that is some, like, shareable assets or anything, I'll definitely be, sharing that as well. But, yeah, any anything you wanna add, Mohawk?\u003C\u002Fp>\u003Cp>Speaker 1: I I think that's all. This was great. I appreciate you, Matt, bringing me on to be able to show the power of Clay and direct us together. And I just hope that, you know, if you're watching this right now, you have the site, you have an idea. Hopefully, the idea juices are flowing right now, and we'll try to make it as easy as possible for you guys to get this kind of up and running.\u003C\u002Fp>\u003Cp>I will say, Matt, to your point, lead to speed is everything now, and I think speed of execution is everything now, right, in the AI age. So, this hopefully will save a lot of time, but also kind of enable you to do things that you were not were not possible before. I think that's the most exciting part for me. But yeah. Again, thank you, Matt.\u003C\u002Fp>\u003Cp>Appreciate the time.\u003C\u002Fp>\u003Cp>Speaker 0: Yeah. Awesome.\u003C\u002Fp>","Hey. What's up, everybody? Matt here from the direct us marketing team. I am super excited today to be showing off, what I think is a really cool session personally. Kinda set the scene a little bit. Obviously, I said marketing, and that's probably terrifying for a lot of people. But, I I like the I I truly believe the gap between, like, technical and nontechnical is getting, you know, smaller. So Mhmm. That's why I'm super excited to have Clay and my friend Mohawk Desai here. Mohawk, how are you, man? Hey. How's it going, Matt? Thanks for bringing me on. Appreciate it. Yeah. Yeah. Absolutely. So most of our audience is probably not familiar with Clay. Yeah. I will say, like, I I think Clay was kind of the first step for me to get to become more technical, with a lot of enrichment and sort of stuff, but like that. But maybe you could give kind of a a brief Yeah. Overview of what Clay is for the folks. Yeah. Absolutely. So at a high level, Clay is a a GTM orchestration and data enrichment tool. And I think a good analogy is I would view Clay as almost a development environment, except not for technical folks, not for coders, but for GTM nerds, like me and like Matt. So, yeah, Clay is if you look at Clay, and we'll jump into the product in a second, but it looks like Excel. There's a lot of AI formulas. There's a lot of, sort sort of technical concepts like webhooks and HTTP APIs. But it is really built for you know, right now, one of our biggest personas is, like, rev ops people at companies, but that's kind of expanding now into marketing folks, into mark ops, and to other personas down the road as well, including, like, sales. But it's really for, like, systems thinkers who wanna build their sort of GTM engine. Clay is kind of the perfect middle ground for all of that, where you can have your entire tech stack flow into Clay. We integrate with 90 plus providers, for data, but as well as sort of GTM execution tools like Gong. And so Clay is kind of like the glue between all your different tools, as well as a place where you could secure data, the things like emails and phone numbers. So, really, it's sort of a one stop shop for GTM orchestration. It's probably the best way I can describe Clay. Yeah. Wow. That's a you're reading off like a prompter there. That was awesome. No. The way that you yeah. The way you said, like, it looks like a spreadsheet when you first get into it, that was, like, my first reaction when I first started testing it. You know, I was like, what? This just looks like another spreadsheet. And then when you start looking at the integrations and the possibilities with, like, enrichment waterfalls, that's what, like, had my holy But speaking of integrations, you know, Directus is, you know, we're we're we've always been kind of a content management system for a lot of folks. So all of the pages for your website, it could be the back end for applications. There's a lot of different use cases. And as we've been getting Clay and Directus closer together, our team has found a lot of really cool use cases that you can really dive into, with that enrichment data. One of which I'm really excited to talk about today. And I feel like was this, like, Clay's, like, bread and butter early on, this sort of, like, personalization type play? I would say not early on. I think, actually, early on in the moat was this the data enrichment piece. On the back end to find emails and phone numbers, Clay takes a waterfall approach where we don't just check one database, but we'll cycle through 90, again, 90 plus providers to find an email or phone number. And so that was the note early on was a higher coverage rate and higher accuracy for data versus under other individual vendors. But, Matt, to your point, I think in the past year or so, a lot of smart rev ops people and marketing people are realizing that Clay's a really good place to do ABM at scale, so account based marketing at scale. And one of them is, yes, starting to create personalized landing pages at scale for your target accounts. But this can also leak into other cool use cases like you can create Google Slides at scale that do the same thing. We have a direct integration of Google Slides, and and so on and so forth. And now in the in the age of AI with all the tools available, we're trying to be a little bit more open to to tools that will allow you to do this. But Directus is, like, a perfect, perfect example of a great integration where we can, really customize our outreach and get those target accounts in our book. Yeah. Yeah. And when you say ABM, I'm sure the the audience is like, another acronym, like marketing and some of the acronyms. But, for folks that don't know, so ABM is is kind of an approach to marketing and sales that, in my opinion, is the way to do it because it's way less intrusive than just, like, shotgun blasting a bunch of people that have no idea who you are. There's a lot of research and, personalization that goes into identifying, you know, 50 accounts, a 100 accounts that are assigned specifically to, like, your go to market team or your reps. So they get to build relationships and really understand, like, the pains and and things within those those companies. And that kinda sets the scene for today, which is, we're gonna do a workflow from beginning to end and and show you how we built it, basically, where it's not like those shitty emails you'll get that's like, hey. Like, I saw you play volleyball, and now let's get on a fifteen minute call. Like, it that's not personalization. Like, personalization is looking at, like, who a company is hiring, you know, what teams are growing, what is probably because of that, some of the struggles they have right now that your product or solution could maybe solve for, and then building out pages and stories around that. But that's hard to do without automation, without things like Clay for the data enrichment. So that's what I'm excited to talk about today. Amazing. Let's do it. Awesome. Alright. So let's hop in. I hope everything is gonna break. I know as soon as I start sharing stuff. It always I don't know why. It always happens. But, let's start with, like, kind of what we're gonna have at the end. That that's kind of our our end goal. We'll start there and work backwards. So, this is just a quick landing page that I spun up, like, ten minutes before this, with clogged code. UIs, you know, you you can spin up a UI really fast now. But the goal is to have this fake AI agency that we just made up called Cottontail, then AI operations team. So they'll handle a lot of the, the manual processes and things like that. So we're gonna have these customized landing pages where it's like, this is the typical personalization you see. Like, all your company name, your, you know, subheader. But what I'm really excited about is focusing in on this section, which is, like, we're gonna use Clay to basically pull out, you know, who are the the number the highest number of jobs, what teams are really growing, and run a couple of, you know, flows on, what does that mean? Like, what are those struggles, the problems they probably have because of that? And then how can our solution fix that? So, basically, a mirror image of what I just said, but this is what it looks like in practice. Any thoughts on, like you know, obviously, working at Clay, like Yeah. Clay, in my opinion, is one of the best marketing. They're I they're like the shining pinnacle right now, everything that y'all do. Mhmm. And I wanna say ABM is a big part of that. Right? Or it could be cool at all. No. Absolutely. As you mentioned, ABM is becoming even a a larger and larger segment of a use case at Clay. I think typically, customers will come in and start with basic enrichment, basic, automated outbound or automated inbound, but those really savvy, you know, companies that know they have sort of a target list of a hundred, two hundred, you know, large companies or or target accounts. ABM is really, really effective at breaking in when everything else doesn't work. And, our CEO, Kareem, has a very cool, analogy or a sort of phrase that encapsulates what ABM should look like, and that's surround sound. You're surround sound of your customers, in in a way that feels so personalized that they they feel obligated to sort of respond or at least curious enough to alright. What do these guys have to say to us? So ABM is like there's a lot that goes into it. One of them is this use case we're gonna jump into, which are these, like, landing pages. Matt, you're correct. This is becoming a very popular thing amongst advanced GTM teams. But outside of just custom landing pages, there's a lot of other things that kind of fold up into ABM. One of them is signals. So taking a signal based approach and having your reps work off intent based data rather than static third party data. So that's one piece. And then the third piece could also be ads. Clay also just launched a new ads tool, but in general, ads is just another way to sort of, again, surround sound, your your personas at these companies. So now imagine a head of marketing at one of your target accounts is getting brilliant, custom landing pages, great emails from your sales teams, are are getting, you know, timely outreach based on some funding events or relevant events to your company, and then you're also seeing ads on LinkedIn and Meta, it's gonna be hard to not be curious about what you do. So that's ABM in a nutshell. And, Matt, I think you're right that that this is picking up a lot in the market. Yeah. Yeah. That's that's awesome. And when you start folding in, like, billboards and buses with your Yes. Exactly. It's inescapable, man. Yes. Inescapable. Exactly. Exactly. But in a, like, in a way, I think it's it's better than just kind of blindfold just going out there, like, blasting you know, pulling a list of a 100,000 contacts and, like, spray and pray method. Spray. Exactly. So, so yeah. So that's what we're gonna do today is really focus on on building these pages out, which I think is really the foundation of it. This is not only something for the customers, but also, you know, for your team members as they start to do the research having, like it's basically like a slide deck in in a lot of ways. So just to kind of show you how I've set up direct us in the back end, so we've got, Cottontail here. I've created two collections. So we've got a landing pages collection. This is where all of the stuff we're gonna do in clay is gonna push, right here. So looking at the actual data model schema for this, it's it's basically everything I showed on that that front end page. So Mhmm. Company name, we're gonna have slug, so this will, like, auto generate the pages when they they come in. Everything is powered by by Railway, which is really the one of the easiest ways to get started with Directus and and build front ends on. We got industry. We got a little testimonial thing. We'll see if we have enough time. I I definitely wanna show how the mini to one, like, data schema stuff works. And then we'll have the personalization content kind of backfilled here. So, like, as developers or technical people are watching this, like, this is how you can knock something out. Like, if you do this for your marketing team, it would be, like, you're you're the hero for for months just giving them something like this. But, also, if you're building applications or products, like, this is a great way to get it into the market and start, you know, marketing the stuff that you're building. But yeah. So that's all built out. Just to show you what it would look like if you were to manually fill this out, yeah, just like a a typical form here. This is just how I have it laid out, but you can do it a lot of different ways. But, but, yeah, I that's really all I had for the director side. Like, as a back end, it's not sexy. It's just like we need to push all this data here, but I'm really excited to show what we've got for Clay. So, Mohawk, I'll actually turn it over to you. Amazing. Thank you, Matt. Great intro. Great context. So now what I'm gonna do is share my screen, and go into Clay. So for folks that have never seen Clay, welcome to Clay. This is the spreadsheet looking tool that you keep hearing about, and so this is what it looks like. So as a starting point for this example, we've uploaded, as our first step, a list of company domains, just their websites. In Clay, though, you can import these company domains from a bunch of other sources. So we have our own company search where you can filter, you know, by, industry, location, revenue count, etcetera. So if you don't already have a list of target accounts in your CRM or in a CSV, you can use Clay to find those target accounts based on your filters. And, as I mentioned, you can upload a CSV as well or connect your CRM and have a live sync between Salesforce or HubSpot, and and Clay. But, again, zooming back out, we are starting with a list of company domains that we have randomly chosen and put into this table. That's the only input we're putting in. That's the I think the key point of all this is that all these other columns you see that have been set up, you set them up once, and anytime you add a new input, it'll automatically run for, for that new input. So, actually, Matt, I I I'm gonna go ahead and sort of enter a new domain here. And, this is my favorite company in the world. So Apple, we're gonna put that a bit second to play. And so I've put this in, and as you can see, it's already autofilled. And I'll kind of now go column by column and break down what we did here to get to that landing page endpoint that Matt showed. So the the first thing is we ran this action called enrich company, and that essentially, in in a JSON array, will return a bunch of data about the company that could be useful. So things like their employee count, their revenue range, their, again, their their, country of origin, things like that. And so the the relevant data points we wanted to pull out for that landing page for each of these companies are really just the company name, the slug, the employee count, and the industry. What we also got, which is pretty cool, and this is kind of taking customization to that, like, final level, is we have URLs to the company's logo as well as a screenshot of their website. So now we've imported their their brand, what they look like, what they feel like, and that can be used if you'd like in your custom landing page, that you're sending to these target accounts. Then from there, we look into jobs. Now jobs in general, I think, are very underrated data point that GTM team should be indexing on because jobs tell you a lot about the company. Tells you what their tech stack is for specific roles. Right? Like, what Mhmm. In that little section that says you should have experience with these technologies, it tells you what they're using. It tells you about their sort of their mission, what they're up to, kind of everything that a candidate would need to know happens to be a lot of the information that we wanna know as sellers and marketers. So, we went ahead and ran an action called find active job openings at this company. And, all all we really needed here was the domain as the input. Again, we're working off of this one input. It's really all we need. And so once you put your domain into this action, that we, you know, we set sort of our filters and we run it, it'll return how many jobs are sort of posted for this as well as the link to each job, which is wonderful because we can then run an AI column on this, active job opening column to kind of analyze those job postings and extract the key insights that we want, like I mentioned. And so we're able to find, which teams are growing fastest, which functions appear to be manual or understaffed. And imagine replacing all of this with the pain points you wanna extract about these companies from their job postings. And so this is this is where we start going into, like, combining the power of AI with the power of of Clay's customization, and then finally that that final layer of direct is coming in as that that content layer. So we've extracted into columns, sort of these different pain points slash operational insights. I shouldn't say pain point, actually. These are specifically how is the team operating, and they're stored in columns, which we'll use to upload that into those, different fields that Matt showed at the beginning. So great. We have our operational insight columns built out. We also use AI to develop, like, personalized headers and sub headers for that landing page. So, essentially, all the inputs we're putting into Directus are all developed in Clay call my call. And so we got personalized sub headers, which are really cool. And then we finally got to those pain points, which I was a little too excited to talk about. But, if we go here to generate, you'll kinda see the prompt, but we're giving it business context. We're giving it, what we do as an agency, as a fake agency. And then it we tell the AI simply what we would tell maybe a new intern or a new hire company, and and we kind of talk to the AI exactly like that, and we get great results. So we're saying, right, three pain points on their operations or go to market team using the inputs from these outputs that we found before. And, for each one kind of return just like a a quick one liner that we can upload to the website. And so this is an example of the final final result here. So that's all the customization we've done in Clay, and the final step to return this data back to Directus is using HTTP API call. Now when I first joined Clay, I'll be honest, I had no idea what this was. And it sounds complicated and scary, but it is a very we try to make it as easy as possible in Clay, and we literally tell you what to copy and paste in here. But this is essentially just a way to share data between the Clay table and Directus. And so in this case, there's a few methods, to get it's it's pretty self explanatory, actually. If you're getting data, that means you're returning data into this table. But in this case, we're posting data, sort of pushing data out. So you selected that as our method. Matt provided me an endpoint from Directus that we can then paste into here. And then, in our body, this was actually, Matt, you wanna Matt actually set this column up. Matt, how did you generate this this sort of body? Yeah. I used AI. So, like I was I was about to say that. Yeah. Perfect. Yeah. No. Yeah. I mean, one of the that this was actually my first time setting this up, and, our, integrations engineer, Lynn Lindsay, has actually built a Directus specific Directus CMS HTTP API. So if you go to, like, set up an enrichment column, just search Directus and it'll pop up, and there's some instructions in there. But I we also have some docs, which I'll show once we hop over from here, of where you can find those. We also talk about how to get it set, set up on our side as well. Yep. Yeah. So the item from get create update. Yeah. Yeah. Exactly. So it's kind of alluding to all the hard work Matt did to set up this column. He's made it super simple and saved this as a public template that you'll be able to access. If you search up direct as HTTP in the clay search bar, you'll see this exact enrichment here, which is the get item from CMS collection. Yeah. Yeah. Great. And, basically, to to to for the, for the JSON, for this, you know, be able to pull this into Directus, what I did when I say I used AI, like, I literally just took a screenshot of the data model from the Directus side that I showed earlier and then just dropped it in a cloud and was like, can you create, like, a JSON placeholder for Glay? And then Beautiful. My my favorite part actually is, like, mapping the actual columns to like, for some reason, it's just like it's like Legos. Like, I just love Yeah. Yeah. Slash then just, like, go through and pick it. Yeah. It's very satisfying. Yeah. That's by the way, Legos for adults or Minecraft for adults is the analogy I use when I think about clay. A big reason why I joined the company. I get to work with Legos every day now. Yeah. Cool. So that that's that's the last step of the Clay workflow. All this data now, because of Matt's hard work with this HTTP API call, flows back in direct us. I will now pass it off back to Matt to show the final product. Yeah. Did, did you run those columns again? All five. Yeah. All all six, actually. So Apple should have run as well. Yeah. Let me see. So I'll you want me to rerun them real quick? Toss it over. Toss the screen over. Okay. I'll just share my screen and see, and we'll just, like Cool. Cut out as necessary. Yeah. Yeah. We'll, like, we'll do it live. Alright. So we have our direct distance here. If I refresh, everything went according to plan. So it looks like Apple got pushed over. So if you, on your side, run that Yeah. HTTP Call call. Gotcha. Again. Yes. So those should come in one by one. And while you're doing that so like I said, all of the data is automatically pushed over. I I didn't have to lift a finger for this. Where it gets interesting now is, like, we have the logo URL, so we can build, you know, like a we have, flows and directives, which I haven't really talked about much, but you can set up, automations from the data inside here. So we can run a flow that's like, hey. Like, take that URL, go and fetch the the logo, and then you'll have, like, you know, all of that to personalize on as well. So, it looks like everything is in here. Yeah. So it looks like all of our other ones got pushed as well. Cool. I mentioned I'll just show you really quick. So if you go to our docs and with that HTTP API that we set up. So we have in here, integrations page. So you can just come in and and search search search Clay, and you should be able to find, yeah, quick and easy setup, how to use, like, the templates and stuff here. And we tell you, everything pretty well. Shout out to Lindsay for for her work on all this. But cool. Alright. So let's see if it works. That's probably Yeah. Probably the most important part of this. Yeah. Alright. So we've got this. We've got the slug, which should auto apply, to the pages. I am gonna pop in. So I I mentioned this earlier, but we have, like the cool thing about the data in Directus is when you create these collections, you can do, like, many to one so you can associate separate collections. So you have a team that just has access to collections with, like, well based access control, and you can have a team that just has access to landing pages. But they can Nice. Their work can, like, counter, like, counteract. So or in a good way, not not in a bad way. But, like, for Apple, right, like, I don't know if we have I guess we can pull a SaaS in. So we would wanna pull in, like, the specific SaaS testimonial here, which you could probably trigger with another flow, but I'm not that advanced. But I also like to do some manual, manual work around this. So let's see. Moment of truth. So coming over to our front end here. Got basic page. Type in apple. There you go. Apple. Beautiful. Ops team. Wow. So we've got our subheader, All of our specific problems where it looks like they're hiring a lot of engineering, so we can for our products, you know, intake and routing workflows, that support the onboarding, that sort of stuff. Yeah. And then our our SaaS testimonial here. So it's like I mean, if if you take out our talking, what? This is probably, like, a two to five minute workflow if you have, like, all of the basic pieces, and you can scale ABN that fast, which with, like, personalization. That's actually good personalization. I like that that's what has always, like, impressed me with this is just how granular you can get and refine and, like, now you have this. Keep iterating, testing, see what resonates. Yes. And I think, Matt, I'll say one thing here. With great power comes great responsibility. I think it's still on the person that's developing the workflow to have good taste. Good taste in what type of personalization do you do versus what you don't. Matt alluded to a great point earlier. Maybe don't mention their kids' Facebook, you know, profile or something random about them that that they're not gonna care about. It sounds contrived. People are tired of getting AI sloppy emails and over personalized messages. So there's a fine line of personalizing tastefully with something like this versus sort of being, having something contrived that's not gonna work. But, nonetheless, the technology is here. It's exciting, and now we're we're really happy to put this in in your hands. Yeah. Yeah. And that's like you just said, like, alluding back to when I said, like, when I said I'm in marketing and people probably cringed and run out, like, the garlic and holy water. Like, we we try to do it the right way. Like, we're we're trying to be there's, like, good marketing and bad marketing. And I tell, like, our engineering team all the time, like, I'm I'm trying to change the way that people think about it. Because a lot of, like Mhmm. A lot of it has been just, like, those quick growth hack things that I hate. But, when you do things like this and you actually start to build relationships and understand and show that, like, you are doing the work to learn about Mayo Clinic and, you know, some of the pains and struggles they have. Like, that's it's not contact level. You don't have to worry about PII. You're scraping all the available data you can find on the site. So, yeah, I I think this is you know, we keep alluding to it, but the way forward and, it's cool to see that Clay and Directus, like, just as a stack can can do it. Yeah. Obviously, AI helping a lot of things, but, I say it all the time. Everybody's becoming a builder, and you're probably seeing this over at Clay too. Like Yep. Them and moat. Like, this used to be, like, for all the developers in this call, probably a pain in the butt because you had marketing asking, like, can you create this page for me? Can you create this state? Like, how do I get this data here? And like I said, I've opened terminal before I worked here at Directus probably twice in my life, and now using cloud code. Being able to do all this myself and not have to bother anybody, that's the way things are shifting for for us as, you know, go to market team. And I'm I'm really excited to see, you know, where we land with it. Absolutely. Well, cool, man. Mohawk, I really appreciate you coming on today, giving us Likewise. You know, a quick view of Clay, how everything works. I'm so glad everything worked out. We didn't have any technical issues on either end. But, for everybody watching, like, all of the prompts that, were shared, I'll make sure that it's in, like, a accessible place. So if you wanna just grab those and test it yourself. Anything else from this call that is some, like, shareable assets or anything, I'll definitely be, sharing that as well. But, yeah, any anything you wanna add, Mohawk? I I think that's all. This was great. I appreciate you, Matt, bringing me on to be able to show the power of Clay and direct us together. And I just hope that, you know, if you're watching this right now, you have the site, you have an idea. Hopefully, the idea juices are flowing right now, and we'll try to make it as easy as possible for you guys to get this kind of up and running. I will say, Matt, to your point, lead to speed is everything now, and I think speed of execution is everything now, right, in the AI age. So, this hopefully will save a lot of time, but also kind of enable you to do things that you were not were not possible before. I think that's the most exciting part for me. But yeah. Again, thank you, Matt. Appreciate the time. Yeah. Awesome.","published",[18,29],{"people_id":19},{"id":20,"first_name":21,"last_name":22,"avatar":23,"bio":24,"links":25},"ca1ac688-ecac-4f25-a4e9-7daf52c8235a","Matt","Minor","b4402ab0-41e4-4fc6-8bf0-769bf39ff114","Director of Demand Generation at Directus",[26],{"url":27,"service":28},"https:\u002F\u002Fdirectus.io\u002Fteam\u002Fmatt-minor","website",{"people_id":30},{"id":31,"first_name":32,"last_name":33,"avatar":34,"bio":35,"links":10},"9bd1432a-b6d8-4dee-9b90-7cb0381e4c59","Mohak","Desal","7bf5276a-1489-4e45-b291-02bd62da629a","Growth Strategy at Clay",[],{"id":38,"number":39,"year":40,"episodes":41,"show":49},"289f6534-7fdd-46df-8c00-89a75469fe41",4,"2026",[42,43,44,45,46,4,47,48],"bbaa3063-6fbe-4d96-bbc7-e50672f9a308","f885409e-0ace-41e5-aca3-faf4dcd7659b","7271f0be-33fd-4cea-b7bd-9c63e74969e1","0baede33-974c-4343-abad-3cea928c8112","a10f99c3-6b45-46e6-b703-64366f150c57","37e28ea2-bec3-40bd-8b09-b9fbb47c1759","68536266-9502-4df8-a295-ef082dfe6fd0",{"title":50,"tile":51},"Leap Week","62816023-fa7e-4a76-b9a1-2733ee2093a6",{"title":10,"meta_description":10},{"id":47,"slug":54,"season":38,"vimeo_id":55,"description":56,"tile":57,"length":58,"resources":10,"people":10,"episode_number":59,"published":12,"title":60,"video_transcript_html":61,"video_transcript_text":62,"content":10,"seo":63,"status":16,"episode_people":64,"recommendations":66},"introducing-builders","1177392929","A new program for the community helping to push Directus forward.\n\n","3ac7b48f-2f4a-4f6c-b1d2-038893bc61e3",21,7,"Introducing: Builders","\u003Cp>Speaker 0: Hello everyone. Welcome to the builder segment as part of Leap Week. I'm Beth. I run community ops here at Directus, and I have the absolute pleasure of talking to you about a brand new community program we launched in the last month called Directus Builders. Builders is a community champion program for people who use Directus, want to share what they're building, and contribute to the community.\u003C\u002Fp>\u003Cp>Whether you're interested in sharing technical insights and receiving amplification from our social channels, joining a network of other DirectUs users, or getting our support for your own community initiatives, this program is for you if you are using directors to build. By joining, you'll enter a private community with other experience builders and our team. It's open to contributors, customers, partners, users, really anyone who uses Directus to build something useful. You don't need to be building something huge, you just need to be building something real. If you're the kind of person who likes helping others figure things out, sharing what you've been learning, or creating something cool, we want to hear from you.\u003C\u002Fp>\u003Cp>Hopefully, that gives you some indication about the types of activities you can be doing as part of directors builders. We've already seen so much enthusiasm from the new joiners, so thank you so much. We have participants from over 13 different countries which is super exciting to see and that is only growing. Next up we have some of our new builders to showcase what they are working on.\u003C\u002Fp>\u003Cp>Speaker 1: Hi. My name is Craig Harmon, and I'm from Perth, Western Australia. I'm a developer, and I've been using for about, five or six years now. And I wanted to show you a project I've been working on, called Nuxtus. Now Nuxtus started life as a boilerplate just for myself to get up and running very quickly with Directus and Nuxt.\u003C\u002Fp>\u003Cp>But it's kind of grown bigger than that now, and it can do automate a lot of processes, for me to get up and running really quickly, with a front end in Nuxt being driven by Directus. So I wanted to show it to you very quickly today. So I'm just on the Nuxt website here. We can see it's very easy to get going. There's an m p x create command.\u003C\u002Fp>\u003Cp>So if we run that in the terminal, what's going to happen now is we're going to run through the standard director setup. So we'll just use local host, give a password. We'll use an SQLite database so we can get up and running nice and quickly. Quickly. And now what's gonna happen is the installer is gonna go away, download and set up Directus just like it normally would if you installed Directus yourself manually.\u003C\u002Fp>\u003Cp>But it's also going to create a client project, with Nuxt and Tailwind and the Directus SDK setup. But above and beyond that, it's also got some other little, packages, which will help us when working specifically with Directus and Nuxt. If you don't wanna use the whole of Nuxtus, you don't have to. Each of the packages are available separately. If you go to the nuxtus.com website and then go to the GitHub page, you'll see that each individual packages can be downloaded and used in your own projects without having to use Nuxtus.\u003C\u002Fp>\u003Cp>But Nuxtus is just a nice, easy, quick way to get a new project up and running. So we just wait for Directus to finish setting up here. Perfect. So we can see now that Directus and Nuxt are all set up. So if we go into our project directory, and if we run npm start, we'll see our front end Nuxt website is ready to go.\u003C\u002Fp>\u003Cp>There we go. Also, what's happened is we've got our Directus set up. If we log in to Directus, this is all now good to go. Now you could stop here. You can start writing your manual code if you wanted to, But Nuxtus also has some other really, nice features that you might wanna take, advantage of.\u003C\u002Fp>\u003Cp>So let's create a new collection in Directus. And let's say, just for the sake of a demo, we're gonna create a blog. Let's use all of the optional fields, and we'll also create a blog title and a main content area for the blog. Great. So we've added them in.\u003C\u002Fp>\u003Cp>That's all ready to go. One more more thing we're gonna do because we want to make this blog available on the front end is we're also gonna need to go to, user settings and let's make the blog publicly readable. And now that we've done that, now now this first demo is not going to be too exciting, but we'll show where we can go from there. So I haven't written any code yet, but if I go to blog, it's going to tell me that I've got no blog articles found because Nuxtus has picked up that I've created a collection called blog, but I haven't put any data in there yet. Now if I go back to my blog content and create an item, This is some content.\u003C\u002Fp>\u003Cp>We've now got some content. Let's go back to our blog and refresh. We're now given the index number of that blog. And if we click it, we get the blog collection name and all of the data about that blog. Okay.\u003C\u002Fp>\u003Cp>Now that's not overly useful, but let's see what's happening in the background. If we bring up our editor. So in server is all of our direct us package. You can see there's a direct us extension here, but we can largely leave that alone. Let's have a look at the client because that's where things get interesting.\u003C\u002Fp>\u003Cp>So we've got a standard Nuxt configuration. We've got Tailwind installed. But if we go into pages, automatically, Nuxtus has created this blog folder with two files in it. The index, which is the list of all the blog articles and also a details page, which at the moment is just echoing out a blog item. But it's more than that.\u003C\u002Fp>\u003Cp>This blog that we've created is now a fully typed object. So it's getting its type information from the director schema. So we can go down here and say, well, we don't want to display the ID of the blog to, list our blog items. We wanna show the title and that's fully typed in TypeScript already. So we save that.\u003C\u002Fp>\u003Cp>We'll see straight away in the background here. We're now using the title to list our blog items and we can go through again. And there's our our blog article that we created. Likewise, in the blog article itself, we don't want to show just the whole object. We want to show the title and then we want to have another div where we show the content.\u003C\u002Fp>\u003Cp>And again, blog is fully typed, so we know what we can get pull out from the object. And there we go. There's our change. So we've now got the name of the collection, the name of the blog article that we're looking at and the content of it and our routing all set up for us with doing very minimal code. So if you're interested in having a look at Nuxtus, please do go to the website, nuxtus.com, and, feel free to have it installed and let me know how you go.\u003C\u002Fp>\u003Cp>Thank you very much.\u003C\u002Fp>\u003Cp>Speaker 2: Today, I would like to show you how to supercharge AI batch processing using Directus MCP with a self learning note taking workflow. Batch processing is really effective for, for situations where you need raw speed, but it kinda breaks down when you need to make creative decisions on individual items and on the batch as a whole. You can't really do that without a human element. AI models, I found, are great for those sorts of tasks, but they break down when you give them a whole bunch of data. And they also struggle with finding the right information to pull in to make those creative decisions.\u003C\u002Fp>\u003Cp>So the solution I found is to teach AI to take notes. By implementing a workflow that lets Directus MCP take notes directly on whatever it's working on, we gain the ability to do things like long running task tracking because the model can pick up where it left off and and, start fresh with a new context window. And we also get things like self improving database access. You take a note on all the different things that you ran into difficulty with and what you did to solve them and how it worked out. And it speeds up future runs because it can read that note and go, okay.\u003C\u002Fp>\u003Cp>So don't do this. Do this, and I'll be able to just continue with the task I was working on. And then by letting it take notes, we can consolidate a whole bunch of data that was stored in the database. Say, for example, you wanted to pull out all of the aliases for your articles that are related to rabbits and you also want the ID for each one. Well, well, you can just dump that in a note and analyze it in future runs.\u003C\u002Fp>\u003Cp>So in our case, Directus is the knowledge backbone. Every note is just a key value pair. I'll go ahead and show you the notes table here. It's a very simple setup. We have a key, which is just a string, and this is what the AI model uses to kind of categorize what the different notes are about.\u003C\u002Fp>\u003Cp>And we have the value, and that's just a big markdown, field. It's a very simple setup, but it's surprisingly powerful. By using semantic keys and explicit instructions on how to record and review notes, we give the AI models a very quick and effective way to find the context they need. So this database has a problem. We have a whole bunch of articles.\u003C\u002Fp>\u003Cp>We have a 124 articles. The titles don't really make any sense. Oh, the bodies are in Latin. And I I can't work with this. So I've created a prompt for Claude to read the article titles and come up with a concept proposal for each one.\u003C\u002Fp>\u003Cp>What could this article actually be about? So let's look at the system prompt here. You can see that there's two things that it's informing the AI model about. That there's an AI notes table in the rough structure of that table and that it needs to read and record its database insights, especially how it solved problems it ran into and just update that note on every run. This gives it the self learning ability we were talking about earlier.\u003C\u002Fp>\u003Cp>It's able to figure out what it ran into last time and fix it on the next run. Now back to the task. I've told the AI model it's a skilled content analyst and writer. We have a huge collection of articles. All the bodies are full of nonsense.\u003C\u002Fp>\u003Cp>All that. Alright. So its job is to record the concepts that it comes up with in an article concepts note. It's not sure what to do with an article based on the title. Be creative.\u003C\u002Fp>\u003Cp>Find a wacky concept. Then we have some very specific instructions. We say always read the template and template article concepts, this is a note, to determine how to structure your output. Always create a new article concepts note for each run. Always give it the exact name.\u003C\u002Fp>\u003Cp>Always process 30 articles. Always review the last note you created before starting to make sure you don't duplicate any work. This template lets us make sure that we get consistent output on every run. Alright. Let's go ahead and run this a few times and see what its output looks like.\u003C\u002Fp>\u003Cp>I turn on auto refresh, and then I'm gonna go into Claude, add a prompt from Directus, and examine articles and propose concepts. If you want to know how to use this feature, go ahead and take a look at the Directus MCP documentation, and then we're just gonna send it. Now one of the fun things you can do with read and write access direct to Directus is if the model makes a mistake, you can ask it to update the prompt to fix its mistake in future runs. Article concepts. It has identified directly that just about every article appears to be lorem ipsum gibberish and occasional test content.\u003C\u002Fp>\u003Cp>Alright. So came up with a bunch of different articles on a bunch of different ideas. In a bit, we're gonna use this to actually write all these different article concepts. But for now, I just wanna keep going through the batch process to show you how it's able to pick up where it left off and continue. On this run, if you notice, it actually picked up that it needed to look up the translations and didn't have to figure that out from the schema this time.\u003C\u002Fp>\u003Cp>Alright, it is a new day. My cloud usage limits have reset, and we are ready to continue with step two, which is generating the articles based on the concepts that we've put together. All of the, article concepts have now been saved in these, concept notes. They take the article title. They try to figure out what on earth, the article should be about and propose a concept.\u003C\u002Fp>\u003Cp>So now we have the next step of the process, which is to generate a actual article for each of the articles we generate concepts on in both English and German. We're gonna go in batches of 10 and see how well that works. We'll take this prompt and it should automatically read over all of our article concept notes and start filling in articles for those. Let's go. So we're gonna use the turn concepts into full fledged articles prompt.\u003C\u002Fp>\u003Cp>That's gonna go and read through all of our article concept notes and start filling in those articles, and I'll show you those articles as it writes them. I have no idea if these articles are gonna make any sense. This is gonna be fun. I'm having fun reading through these articles. I have no idea how helpful any of this is, but it at least sounds very convincing.\u003C\u002Fp>\u003Cp>Now the key advantage to this approach is that if we were generating concepts and writing articles at the same time, we'd have to use much smaller batches. But because we split the process into two steps, we've saved a ton of context window, and we're able to work in larger batches. And as a bonus, we can do multiple things with those article summaries, while we're working on generating those articles from those summaries. Now this isn't the most efficient workflow for generating articles. There's way better workflows for that.\u003C\u002Fp>\u003Cp>We're just demonstrating this concept. But a nice thing about this approach is that you can perform multiple tasks at the same time. For example, while we were working on these articles, we could also be working on something completely different using the notes that we're generating the articles from. I have this prompt here, categorize article concepts, which will allow Claude to suggest an article taxonomy based on the summaries it's already generated. It'll just read all the article concept notes and create a new note suggesting that new taxonomy.\u003C\u002Fp>\u003Cp>I like the way it describes these articles as a fascinating collection of technical and business focused article concepts with creative jargon filled titles that have been transformed into practical valuable content ideas. I'm not quite sure I'd be so positive about it, but Nate gets the idea across. Alright. Let's take a look at the categories that it generated and the opportunities that identified. So here's some content gaps, AI and ethics, sustainable technology operations, human centered digital transformation, all sorts of other different things.\u003C\u002Fp>\u003Cp>And here are some of the categories it's come up with. System architecture, it's put a bunch of different articles in that category, ecommerce, digital business, data management and analytics, user experience and interface design, business operations or business optimization management, automation and AI systems, project management and collaboration. So it's basically taken all of those articles that we put together and categorized them according to common themes in those articles. This could be really useful if you have an existing content collection that you're trying to build a new categorization system for. And what's great is you could just take this and create another prompt to actually apply that taxonomy to articles, and create categories and all sorts of different things for that in Directus.\u003C\u002Fp>\u003Cp>So by the end of this run, here's what the AI model has built. Concept proposals for every article in our database, full body content for many of those articles in English and German, a whole new content categorization system for the articles it wrote, progress logs so that it doesn't lose its place, and database access hints that documents how the challenges were solved so that on future runs, it could do even better. Instead of treating AI like a disposable worker or one that can infinitely stuff and stuff and stuff and work on something until it gets all confused, We're treating it like a teammate that tracks and follows up on what it's done across multiple days, multiple runs, starting with a clean slate on each task and only pulling in the information that it needs. We turned a human in the loop batch processing task into an AI creative workflow by splitting it into discrete steps and letting it kind of pick up where it left off. We gave the model a way to track progress.\u003C\u002Fp>\u003Cp>We built templates to make the results more consistent and reusable, and we recorded information about the database to help the AI model get smarter over time. I've had a ton of fun nailing down this workflow and testing it on all sorts of tasks. For my work, we're actually using it to do things similar to this where we're analyzing a huge amount of content and trying to figure out how to organize and categorize and and do all sorts of things, and it's it's doing really well. The ability to take notes, record what it's done, and just kind of start over and create its own context has been incredibly helpful with turning a workflow that works really well for one or two articles into something that works well across thousands. So So thank you for your time.\u003C\u002Fp>\u003Cp>I really look forward to seeing what everyone does with Directus and Directus MCP going forward, especially as Directus MCP evolves, AI models get more capable, and just the overall core gets stronger. The future is gonna be fun.\u003C\u002Fp>\u003Cp>Speaker 3: Hey. I'm Matt from Morley. We're a directors partner agency based in The UK. And today, I want to give you a quick look at what we've been building, both for our clients and for the directors community. So one of the things we built for directors that we're most proud of is the NUXT directors module.\u003C\u002Fp>\u003Cp>It's an open source NUXT module that wraps the official directors SDK and gives you proper first class integration with NUXT. The idea is simple, you add your module, point it at your Directus instance and that's it, you're up and running. You get session based authentication with cross domain support out of the box, automatic type generation from your directors collections, type safe websockets, Nux images automatically configured for your directors assets, and we even embed the directors admin panel right into the Nux dev tools. So let me show you how to get up and running with that. So I'll just open up Visual Studio Code.\u003C\u002Fp>\u003Cp>The first thing you need to do is go into your Nux config, add the Nux Direct SDK, install it in whatever package manager you like, then go to your environment variable, add your direct to your URL, and if you want to use things like, admin server endpoints or automatic type generation then just add in a Nux, a direct to admin token as well and that will be automatically handled for you. Don't worry nothing will be passed to the front end it's all completely safe. And that's it, when you start the server you get access to everything you'd expect from your Direct SDK automatically imported. So as you can see here on my blog page I've got the directors all I do is call use directors and then things like read items automatically imported and it's fully type safe. You can see all of my collections.\u003C\u002Fp>\u003Cp>I get errors if it's a if it doesn't actually exist in the director's schema and I can do things that you'd expect. Fully typed filtering, fully typed nested fields, and again I will get errors if something doesn't exist, and fully typed sort. That's it. It's up, it's running. You can do, there's even help methods for things like getting your file URLs, where you can pass in options around your thumbnails, whether you want to download it, what you want to call it, all that is handled automatically for the NUXT Directus SDK.\u003C\u002Fp>\u003Cp>We built this because we kept solving the same problems across client projects and rather than copying and pasting boilerplate we packaged it up properly and made available on NPM. It's MIT licensed and we're just about to release version five which targets Nux four and the latest Direct SDK. Here's some of the work that we built using Directus and the Nux Directus SDK. So that's a quick snapshot of what we're building at Rolla. If you're building anything with Nux and Directus, feel free to give the Nux Directus SDK a look.\u003C\u002Fp>\u003Cp>Otherwise, if you want to chat about anything Nux, directors or even potentially a future project please do reach out at rolli. Io. Thank you very much.\u003C\u002Fp>\u003Cp>Speaker 0: Thank Thank you so much for the showcases and to everyone who's already joined and showed enthusiasm. We are really excited about bringing the community together in this way and to see all the possibilities that can come out of it and, yeah, hopefully, we have done enough to convince you to join. That's it from me. If you have any questions, do head over to the community forum at community.directors.io and have a great rest of leap week, everyone. See you soon.\u003C\u002Fp>\u003Cp>Bye.\u003C\u002Fp>","Hello everyone. Welcome to the builder segment as part of Leap Week. I'm Beth. I run community ops here at Directus, and I have the absolute pleasure of talking to you about a brand new community program we launched in the last month called Directus Builders. Builders is a community champion program for people who use Directus, want to share what they're building, and contribute to the community. Whether you're interested in sharing technical insights and receiving amplification from our social channels, joining a network of other DirectUs users, or getting our support for your own community initiatives, this program is for you if you are using directors to build. By joining, you'll enter a private community with other experience builders and our team. It's open to contributors, customers, partners, users, really anyone who uses Directus to build something useful. You don't need to be building something huge, you just need to be building something real. If you're the kind of person who likes helping others figure things out, sharing what you've been learning, or creating something cool, we want to hear from you. Hopefully, that gives you some indication about the types of activities you can be doing as part of directors builders. We've already seen so much enthusiasm from the new joiners, so thank you so much. We have participants from over 13 different countries which is super exciting to see and that is only growing. Next up we have some of our new builders to showcase what they are working on. Hi. My name is Craig Harmon, and I'm from Perth, Western Australia. I'm a developer, and I've been using for about, five or six years now. And I wanted to show you a project I've been working on, called Nuxtus. Now Nuxtus started life as a boilerplate just for myself to get up and running very quickly with Directus and Nuxt. But it's kind of grown bigger than that now, and it can do automate a lot of processes, for me to get up and running really quickly, with a front end in Nuxt being driven by Directus. So I wanted to show it to you very quickly today. So I'm just on the Nuxt website here. We can see it's very easy to get going. There's an m p x create command. So if we run that in the terminal, what's going to happen now is we're going to run through the standard director setup. So we'll just use local host, give a password. We'll use an SQLite database so we can get up and running nice and quickly. Quickly. And now what's gonna happen is the installer is gonna go away, download and set up Directus just like it normally would if you installed Directus yourself manually. But it's also going to create a client project, with Nuxt and Tailwind and the Directus SDK setup. But above and beyond that, it's also got some other little, packages, which will help us when working specifically with Directus and Nuxt. If you don't wanna use the whole of Nuxtus, you don't have to. Each of the packages are available separately. If you go to the nuxtus.com website and then go to the GitHub page, you'll see that each individual packages can be downloaded and used in your own projects without having to use Nuxtus. But Nuxtus is just a nice, easy, quick way to get a new project up and running. So we just wait for Directus to finish setting up here. Perfect. So we can see now that Directus and Nuxt are all set up. So if we go into our project directory, and if we run npm start, we'll see our front end Nuxt website is ready to go. There we go. Also, what's happened is we've got our Directus set up. If we log in to Directus, this is all now good to go. Now you could stop here. You can start writing your manual code if you wanted to, But Nuxtus also has some other really, nice features that you might wanna take, advantage of. So let's create a new collection in Directus. And let's say, just for the sake of a demo, we're gonna create a blog. Let's use all of the optional fields, and we'll also create a blog title and a main content area for the blog. Great. So we've added them in. That's all ready to go. One more more thing we're gonna do because we want to make this blog available on the front end is we're also gonna need to go to, user settings and let's make the blog publicly readable. And now that we've done that, now now this first demo is not going to be too exciting, but we'll show where we can go from there. So I haven't written any code yet, but if I go to blog, it's going to tell me that I've got no blog articles found because Nuxtus has picked up that I've created a collection called blog, but I haven't put any data in there yet. Now if I go back to my blog content and create an item, This is some content. We've now got some content. Let's go back to our blog and refresh. We're now given the index number of that blog. And if we click it, we get the blog collection name and all of the data about that blog. Okay. Now that's not overly useful, but let's see what's happening in the background. If we bring up our editor. So in server is all of our direct us package. You can see there's a direct us extension here, but we can largely leave that alone. Let's have a look at the client because that's where things get interesting. So we've got a standard Nuxt configuration. We've got Tailwind installed. But if we go into pages, automatically, Nuxtus has created this blog folder with two files in it. The index, which is the list of all the blog articles and also a details page, which at the moment is just echoing out a blog item. But it's more than that. This blog that we've created is now a fully typed object. So it's getting its type information from the director schema. So we can go down here and say, well, we don't want to display the ID of the blog to, list our blog items. We wanna show the title and that's fully typed in TypeScript already. So we save that. We'll see straight away in the background here. We're now using the title to list our blog items and we can go through again. And there's our our blog article that we created. Likewise, in the blog article itself, we don't want to show just the whole object. We want to show the title and then we want to have another div where we show the content. And again, blog is fully typed, so we know what we can get pull out from the object. And there we go. There's our change. So we've now got the name of the collection, the name of the blog article that we're looking at and the content of it and our routing all set up for us with doing very minimal code. So if you're interested in having a look at Nuxtus, please do go to the website, nuxtus.com, and, feel free to have it installed and let me know how you go. Thank you very much. Today, I would like to show you how to supercharge AI batch processing using Directus MCP with a self learning note taking workflow. Batch processing is really effective for, for situations where you need raw speed, but it kinda breaks down when you need to make creative decisions on individual items and on the batch as a whole. You can't really do that without a human element. AI models, I found, are great for those sorts of tasks, but they break down when you give them a whole bunch of data. And they also struggle with finding the right information to pull in to make those creative decisions. So the solution I found is to teach AI to take notes. By implementing a workflow that lets Directus MCP take notes directly on whatever it's working on, we gain the ability to do things like long running task tracking because the model can pick up where it left off and and, start fresh with a new context window. And we also get things like self improving database access. You take a note on all the different things that you ran into difficulty with and what you did to solve them and how it worked out. And it speeds up future runs because it can read that note and go, okay. So don't do this. Do this, and I'll be able to just continue with the task I was working on. And then by letting it take notes, we can consolidate a whole bunch of data that was stored in the database. Say, for example, you wanted to pull out all of the aliases for your articles that are related to rabbits and you also want the ID for each one. Well, well, you can just dump that in a note and analyze it in future runs. So in our case, Directus is the knowledge backbone. Every note is just a key value pair. I'll go ahead and show you the notes table here. It's a very simple setup. We have a key, which is just a string, and this is what the AI model uses to kind of categorize what the different notes are about. And we have the value, and that's just a big markdown, field. It's a very simple setup, but it's surprisingly powerful. By using semantic keys and explicit instructions on how to record and review notes, we give the AI models a very quick and effective way to find the context they need. So this database has a problem. We have a whole bunch of articles. We have a 124 articles. The titles don't really make any sense. Oh, the bodies are in Latin. And I I can't work with this. So I've created a prompt for Claude to read the article titles and come up with a concept proposal for each one. What could this article actually be about? So let's look at the system prompt here. You can see that there's two things that it's informing the AI model about. That there's an AI notes table in the rough structure of that table and that it needs to read and record its database insights, especially how it solved problems it ran into and just update that note on every run. This gives it the self learning ability we were talking about earlier. It's able to figure out what it ran into last time and fix it on the next run. Now back to the task. I've told the AI model it's a skilled content analyst and writer. We have a huge collection of articles. All the bodies are full of nonsense. All that. Alright. So its job is to record the concepts that it comes up with in an article concepts note. It's not sure what to do with an article based on the title. Be creative. Find a wacky concept. Then we have some very specific instructions. We say always read the template and template article concepts, this is a note, to determine how to structure your output. Always create a new article concepts note for each run. Always give it the exact name. Always process 30 articles. Always review the last note you created before starting to make sure you don't duplicate any work. This template lets us make sure that we get consistent output on every run. Alright. Let's go ahead and run this a few times and see what its output looks like. I turn on auto refresh, and then I'm gonna go into Claude, add a prompt from Directus, and examine articles and propose concepts. If you want to know how to use this feature, go ahead and take a look at the Directus MCP documentation, and then we're just gonna send it. Now one of the fun things you can do with read and write access direct to Directus is if the model makes a mistake, you can ask it to update the prompt to fix its mistake in future runs. Article concepts. It has identified directly that just about every article appears to be lorem ipsum gibberish and occasional test content. Alright. So came up with a bunch of different articles on a bunch of different ideas. In a bit, we're gonna use this to actually write all these different article concepts. But for now, I just wanna keep going through the batch process to show you how it's able to pick up where it left off and continue. On this run, if you notice, it actually picked up that it needed to look up the translations and didn't have to figure that out from the schema this time. Alright, it is a new day. My cloud usage limits have reset, and we are ready to continue with step two, which is generating the articles based on the concepts that we've put together. All of the, article concepts have now been saved in these, concept notes. They take the article title. They try to figure out what on earth, the article should be about and propose a concept. So now we have the next step of the process, which is to generate a actual article for each of the articles we generate concepts on in both English and German. We're gonna go in batches of 10 and see how well that works. We'll take this prompt and it should automatically read over all of our article concept notes and start filling in articles for those. Let's go. So we're gonna use the turn concepts into full fledged articles prompt. That's gonna go and read through all of our article concept notes and start filling in those articles, and I'll show you those articles as it writes them. I have no idea if these articles are gonna make any sense. This is gonna be fun. I'm having fun reading through these articles. I have no idea how helpful any of this is, but it at least sounds very convincing. Now the key advantage to this approach is that if we were generating concepts and writing articles at the same time, we'd have to use much smaller batches. But because we split the process into two steps, we've saved a ton of context window, and we're able to work in larger batches. And as a bonus, we can do multiple things with those article summaries, while we're working on generating those articles from those summaries. Now this isn't the most efficient workflow for generating articles. There's way better workflows for that. We're just demonstrating this concept. But a nice thing about this approach is that you can perform multiple tasks at the same time. For example, while we were working on these articles, we could also be working on something completely different using the notes that we're generating the articles from. I have this prompt here, categorize article concepts, which will allow Claude to suggest an article taxonomy based on the summaries it's already generated. It'll just read all the article concept notes and create a new note suggesting that new taxonomy. I like the way it describes these articles as a fascinating collection of technical and business focused article concepts with creative jargon filled titles that have been transformed into practical valuable content ideas. I'm not quite sure I'd be so positive about it, but Nate gets the idea across. Alright. Let's take a look at the categories that it generated and the opportunities that identified. So here's some content gaps, AI and ethics, sustainable technology operations, human centered digital transformation, all sorts of other different things. And here are some of the categories it's come up with. System architecture, it's put a bunch of different articles in that category, ecommerce, digital business, data management and analytics, user experience and interface design, business operations or business optimization management, automation and AI systems, project management and collaboration. So it's basically taken all of those articles that we put together and categorized them according to common themes in those articles. This could be really useful if you have an existing content collection that you're trying to build a new categorization system for. And what's great is you could just take this and create another prompt to actually apply that taxonomy to articles, and create categories and all sorts of different things for that in Directus. So by the end of this run, here's what the AI model has built. Concept proposals for every article in our database, full body content for many of those articles in English and German, a whole new content categorization system for the articles it wrote, progress logs so that it doesn't lose its place, and database access hints that documents how the challenges were solved so that on future runs, it could do even better. Instead of treating AI like a disposable worker or one that can infinitely stuff and stuff and stuff and work on something until it gets all confused, We're treating it like a teammate that tracks and follows up on what it's done across multiple days, multiple runs, starting with a clean slate on each task and only pulling in the information that it needs. We turned a human in the loop batch processing task into an AI creative workflow by splitting it into discrete steps and letting it kind of pick up where it left off. We gave the model a way to track progress. We built templates to make the results more consistent and reusable, and we recorded information about the database to help the AI model get smarter over time. I've had a ton of fun nailing down this workflow and testing it on all sorts of tasks. For my work, we're actually using it to do things similar to this where we're analyzing a huge amount of content and trying to figure out how to organize and categorize and and do all sorts of things, and it's it's doing really well. The ability to take notes, record what it's done, and just kind of start over and create its own context has been incredibly helpful with turning a workflow that works really well for one or two articles into something that works well across thousands. So So thank you for your time. I really look forward to seeing what everyone does with Directus and Directus MCP going forward, especially as Directus MCP evolves, AI models get more capable, and just the overall core gets stronger. The future is gonna be fun. Hey. I'm Matt from Morley. We're a directors partner agency based in The UK. And today, I want to give you a quick look at what we've been building, both for our clients and for the directors community. So one of the things we built for directors that we're most proud of is the NUXT directors module. It's an open source NUXT module that wraps the official directors SDK and gives you proper first class integration with NUXT. The idea is simple, you add your module, point it at your Directus instance and that's it, you're up and running. You get session based authentication with cross domain support out of the box, automatic type generation from your directors collections, type safe websockets, Nux images automatically configured for your directors assets, and we even embed the directors admin panel right into the Nux dev tools. So let me show you how to get up and running with that. So I'll just open up Visual Studio Code. The first thing you need to do is go into your Nux config, add the Nux Direct SDK, install it in whatever package manager you like, then go to your environment variable, add your direct to your URL, and if you want to use things like, admin server endpoints or automatic type generation then just add in a Nux, a direct to admin token as well and that will be automatically handled for you. Don't worry nothing will be passed to the front end it's all completely safe. And that's it, when you start the server you get access to everything you'd expect from your Direct SDK automatically imported. So as you can see here on my blog page I've got the directors all I do is call use directors and then things like read items automatically imported and it's fully type safe. You can see all of my collections. I get errors if it's a if it doesn't actually exist in the director's schema and I can do things that you'd expect. Fully typed filtering, fully typed nested fields, and again I will get errors if something doesn't exist, and fully typed sort. That's it. It's up, it's running. You can do, there's even help methods for things like getting your file URLs, where you can pass in options around your thumbnails, whether you want to download it, what you want to call it, all that is handled automatically for the NUXT Directus SDK. We built this because we kept solving the same problems across client projects and rather than copying and pasting boilerplate we packaged it up properly and made available on NPM. It's MIT licensed and we're just about to release version five which targets Nux four and the latest Direct SDK. Here's some of the work that we built using Directus and the Nux Directus SDK. So that's a quick snapshot of what we're building at Rolla. If you're building anything with Nux and Directus, feel free to give the Nux Directus SDK a look. Otherwise, if you want to chat about anything Nux, directors or even potentially a future project please do reach out at rolli. Io. Thank you very much. Thank Thank you so much for the showcases and to everyone who's already joined and showed enthusiasm. We are really excited about bringing the community together in this way and to see all the possibilities that can come out of it and, yeah, hopefully, we have done enough to convince you to join. That's it from me. If you have any questions, do head over to the community forum at community.directors.io and have a great rest of leap week, everyone. See you soon. Bye.","aed3883f-f0df-4ae7-8c85-5df7e2c86161",[65],"912b46f7-c6c2-4876-a81c-1060ff910431",[],1781213219382]