Build Executive Slide Decks in Claude Code with AI — Data Neighbor Live

0:00 Everybody, thanks for joining in. Today's lesson is on how to build slide decks in cloud code. And hopefully um hopefully this is uh uh uh sort of like something interesting for you and um uh by the end um hopefully you'll be able to at least see the capability of what cloud code can do on things beyond just coding. Um so just a quick intro for all three of us. Um, my name is Hi. I am the head of data at a legal tech company called Entra. Um, I was previously at other consumer tech companies um at Nextoor, LinkedIn, Pinterest, Meta, so on and so forth. So been in the data science analytics space for uh almost two or almost 20 years uh now and uh yeah excited to to lead this session and uh talk to you guys about this. Um I will hand it off to Sean or Shravia pass it to Travia.

1:03 >> Sure. Hey everyone, I'm Stravia. I'm a data sense leader with experience in Microsoft eBay next door. That's where I met Sean and hi and uh most recently at superhuman. Yeah. Uh very excited to share everything we are doing in the analytics and AI space cloud code uh and you know we also started working to look into codeex and other things. So excited to share with you all our learnings as part of these workshops. Hand over to Sean.

1:35 >> Hey everyone, I'm Sean. Uh yeah lead AI analyst lab with high and stravia. uh been in data science for about 10 years. Currently a principal data scientist at a same [clears throat] legal tech company that high works at and I specialize primarily in a gentic analytics and uh AI evaluation. But uh today we're going to be talking about slide decks. So pass that off to high. >> Cool. Yeah, we're going to talk about slide decks uh without touching PowerPoint. So hopefully um this would be something interesting and uh uh you'll walk away with a little bit of mental model for how to do that yourself.

2:15 Okay, cool. Let's see. Oh, and and just before we get into the rest of the content, um so we're all from AI Analyst Lab. You can check us out on AIan Analystlab.ai. We have a lot of free courses, free workshops of the past that you can watch recordings for. We also have a lot of upcoming uh workshops and lessons just like this. So you can sign up for those as well. We run a couple boot camps or I guess weekend workshops um that you can join us as well that we get more deeper into uh cloud code uh in terms of building AI analyst systems and uh uh and uh and setting up quad code and things like that. We'll get into that a little bit more as we go.

3:01 Cool. Okay. So, here's uh here's here's the scenario that we're gonna uh that we're going to walk through. Um so, suppose you have the data of what whatever you're analyzing. So, you know, we're all data professionals here. Um the three of us, uh we typically have to do a lot of um kind of like presentations and uh uh deck buildings to convey our findings from the data that we have. And so in the upcoming uh you know like uh rest of the lesson we're going to ground it in kind of like the data uh analogy or the the the data world uh sort of and so you know like we do or maybe many of you do as well do a lot of like data analysis or some sort of information gathering and uh you have a deck that's due tomorrow and most of the time um I guess in the old way uh for the most card. It's a, you know, spend a lot of time in either PowerPoint or in Google Slides. Uh, and maybe there's some templates that you already have that matches your company's themes and brands and things like that, but like you still have to like drag and drop and do all of that kind of stuff uh across the board. Now with AI tools and uh uh just you know generic general sort of like horizontal uh chat bots um or other agents that can help um things like cloud code work and stuff like that. Uh you now have a lot better control and a lot better tooling to help you do this where you know perhaps you can fire off a prompt and then you can uh you can get some uh some stuff back uh pretty quickly. Uh so you know like things have advanced really really fast and uh um and there are also like other uh tool building or presentation building sort of tools out there like gamma um so on and so forth. What we're talking about today here is to leverage clock code to do the end to end such that you can uh like the final output of a deck could actually look um look like something that's been reviewed by you know like multiple kind of like experts and uh things are very coherent in terms of the storytelling aspect of it.

5:18 So, um, I just pulled these random screens from, uh, from kind of like a, uh, uh, using claude co-work to prompt for a presentation tech. So, many of you might have heard of the term AI slop, right? Like, uh, or I guess maybe let me let me ask you this. Uh, drop in chat if you if you feel like this these slides look great.

5:50 Okay. Claire says no. Stravia says not bad. W says they look okayish. Okay. Seen worse for sure. Okay. Wow. Looks clawed. Not great. They don't feel like real world decks I would put together at work. Average. Yeah. So, they're pretty mid, right? Like, it's looks pretty generic. It's got the themes that feels like it comes from an AI. And uh for the most part, well, I mean, like, you know, to be fair, I I I didn't know what the topic was on this one. Apparently, it's on it's for octopus. Uh but, you know, I don't know how interesting you can make octopus. Um but just looking at the the look and feel of uh of of the slides here, uh it doesn't look, you know, like super compelling, right? even if you expand it to the entire deck, it's probably going to be something very similar for the rest of the of the sequence, if you will. So um so the reason why kind of like you know if you don't give it enough context or instructions that AI defaults to something like this is pretty simple because um it's trying to just do the fastest way of generating accomplishing the goal that you set out that you set it out to do. Um I gave it a pretty generic prompt of uh hey build me a deck about uh a random topic and then that's what it gave me. And so um the power a lot of times in being able to leverage tools like claude or cloud code or uh chat gbt is the ability to give it instructions. And so we're going to go through what a wellchained kind of like um sequence of instructions would would uh would bring in terms of the output and the uh and the difference of it.

7:44 And if anyone has any questions um along the way, feel free to drop them in chat and we'll answer them or uh you know raise your hands and uh we'll we'll uh we'll have a Q&A at the uh at the end. Cool. Okay. So, um what we're going to be talking about is uh all of you might have seen might have seen our LinkedIn posts uh of the past or have been at uh uh past lightning lessons that we've run. Um we're going to be using the we're going to show the the uh AI analyst uh repo that we've uh uh open sourced and uh uh created for everyone to freely download and uh work with it.

8:29 Basically what this is is an agentic uh system that you can work with in cloud code that basically mimics the uh the uh the ability of a senior product data scientist. And so um you know like feel free to feel free to clone it. Feel free to uh to take a look at it. It's got many different agents skills that effectively is trying to do something that a product data scientist would be doing and all the best practices that go into it. So it uh it it it understands how to frame better questions. It understands how to uh uh how to how to how to define metrics. Uh understands how to um provide guard rails for the metrics. Understands how to do uh experimentation. And then obviously the final um piece of it is like um making recommendations uh having getting the context about the business and then making connections of them and uh building decks and stuff like that. So, we're going to take a look at the deck building component of uh of this full pipeline, and we'll go into some of the uh uh some of what it does, and then have it sort of show us uh what what it can do at the end.

9:43 Well, uh let's see. Uh oh, and uh if Shavei or Shan if you want to drop the link to this guy um so people can clone it, that would be, you know, folks feel free to follow along as well. >> Yep. Drop it in the chat. >> Okay, cool. Great. So, let's see. Um cool. Okay. Um so, you know, like we have the we have the Aentic system freely available. Um, we also run different workshops and boot camps where you can you also can get to kind of like learn it with us in terms of both how to set it up, how to run against it, and also how to how to set up something very similar uh for yourself. Like if you want to build something from scratch, we have a boot camp that teaches folks how to how to actually do it. We've run one in uh back in April and uh um the the sort of like the vibe was really really really really cool and uh I think our students got a lot out of it. Um we'll share more about uh the specifics later down the uh later towards the end of the of the lesson. Um uh but just keep in mind that uh we have these things um uh that we run these things as well. And so if you're interested u we'd love to have you join us. Uh, and perhaps Sean Dravia, you guys can drop a link to kind of like those things for people to look at as well, but we'll come back to it.

11:12 Okay, cool. So, um, so what I wanted to get into is kind of like the few components that uh that makes uh decks look better than for example like if you just prompt AI out of the box uh for uh for for a deck. So, um, the four things that are pretty important to to nail for AI to have enough information to get a really good looking and compelling deck are these things and would love to know if people feel like, you know, more should be more should be included or like uh other things that you've uh that you've seen that works. Number one is obviously the audience like telling or understanding who you're presenting to and having that context available is pretty important. Is it an an executive readout? Is it a workshop like this one?

12:10 Is it a team standup? uh or you know like what whatever the setting is really gives the highle overarching kind of uh context setting that's really important for both the tone but also like the information architecture for what should be included in the deck itself. This is just like humans, right? Like this is uh you know like we always think about who the audience is and then tailor the message based on that. If it's for you know like in the world of data if it's for like the data scientists you can go probably go pretty technical on the on the deck itself and explain methodologies and findings and whatever it is in a very detailed manner and show your work very extensively. But if it's for for example like a CEO that would be completely different. you're probably going to mask away 90% of all those things and then just show the takeaways, the important evidence and what you're asking for. So, and then every everything else it would be somewhere in between. So it's uh you know like audience is the most important piece where having that context helps really shape the uh the you know even how many slides we should have in the uh in the deck itself versus uh uh you know some other some other audience that would be caring about something completely different. So that's number one. Number two is the context um specifically the decision that the deck is driving. So for example here is um you know like hey if I have a deck that conveys uh you know like a like a topic around should we be shipping a specific feature then that's a very different a very different messaging than if uh we're producing a deck for are we on track of like you know um progress updates and then that would be very different from if you're asking for resources to further some sort of investment or uh or uh uh go double down on some sort of uh some sort of initiatives. So the decision is also really important here where you know like in addition to the audience is the intent of what you're trying to do with the deck.

14:29 Make sense so far? Hopefully. And then once you have those then you need a story right like um you know hey what is the narrative uh so uh right here it says storyboard before slides. So you know like typically there's a structure for what works um in terms of the flow of how you're going to deliver the presentation. So it might have something like a TLDDR might have the context setting so that everyone's on the same page before you get into the detail with certainly for the case of analysis um and presenting findings out of it. Obviously the actual findings that you saw that are most relevant for the group of people that you're presenting to that's really really important. That's the meat of it. And then also like you know the recommendation like so what um you know having gone through the actual presentation or the slide deck uh what are you trying to accomplish at the end of the day? So there's always like there's different components that make up depending on the audience again um that make up your sequence of messaging and what gets accomplished at the end of the day. But that's all can be wrapped into this whole category of uh something like a narrative arc. It's like, you know, like at the at the end of the day, what is it? And it's not a one-sizefits-all. Um, it certainly is, you know, not a chart dump in chronological order. I don't know if a lot of you have seen some, you know, like let's call it like a a lot to be desired s sort of presentations where it's just one chart after another with very generic titles and at the end you're like, okay, then like what what what is what is going on here? What do you want me to take away? um and you know left scratching their heads. So what we're trying to do here is uh hopefully steer away from that sort of output and then we'll see how AI can do it uh itself.

16:29 And then finally here's the uh more more kind of like the um uh the userfacing aspect of it. So what is the theme of the uh of the of the presentation? Theme meaning like literally like what what's the look and feel? Um so you know perhaps uh and and we'll show a few themes uh as well in the demo. Uh but uh you know like if something is more fun or whatever you know you can have a different theme than uh if it's something more formal for example.

17:00 There's a reason why uh uh what do you call this like uh McKenzies or the consulting companies are very particular about uh what they want in their uh in their client decks and things like that precisely because they want to make sure that uh number one it's being consistent number two it reflects kind of like the um the nature of uh of of the uh of the talks that or the the presentations that they're giving. So, you know, like I think if if it's uh uh hopefully no no surprise that uh you know, all these things are actually carried over to the uh the AI analyst repo that we just talked about and uh it knows how to do sort of like all of these uh independently like by itself.

17:47 So, um the we call it like the pipeline. So basically it's like a um you know like a sequence of steps for the AI to execute um all the skills and agents uh effectively tries to orchestrate a uh kind of like a sequence of events or a sequence of uh execution steps that leads to the final output and it mimics what we just talked about in terms of the sequence. So it's got the storyboard where you know like given a set of data given a set of findings. So assume that we've done the analysis up front because this is a building a deck sort of course or lesson here. Uh we're going to you know not look at the actual analysis or how that's done and stuff like that. So in your mental model assume that we have uh a set of completed analysis and now we need to build a deck for it. And so the pipeline runs like okay now you have a bunch of uh stuff. So let's see what makes sense to build a uh to build a storyboard. So what's the arc? What is the uh what is the climax? What is the uh what is the information that we should be telling people and uh kind of like plan it high level like that. And then the second piece is uh the narrative like how should [snorts] I tell that story based on the story arc based on the storyboard and based on the actual data and finding that it's already been done. Uh so it understands how to do executive summary, how to do which findings to pick out, what recommendations you can do um based on those and then connect it back to sort of like the business context or whatever else it has access to to formulate the actual um you know like the the actual messaging on the de on the slides themselves. And then they'll build the charts. And then you know compiling all that together becomes the deck. And then the fifth the fifth u the fifth step here is where humans come in to you know like hey maybe play around with it give it feedback and uh have it iterate. Um and we'll see all of that in just a little bit. Does this make sense? like generally speaking in terms of a flow like hopefully this is similar to how you would think about doing it yourself if you were to build it manually right >> yeah I think I think one of the the other things about this is that uh you know you can do this for any use case it doesn't have to be analytics but in terms of like analysis which is obviously our kind of focus area so much of the work goes into cleaning the data, framing the questions, doing the analysis, reviewing the analysis, like just, you know, so much work that it's really easy at the end of the day to kind of like hand off something that's not in a very well curated deck that doesn't have a story or a narrative to it. And it's just a shame when that happens because so many so much like hard work that really could be driving decisions doesn't end up being actioned on because the story or the deck like doesn't get um showcased or surfaced um in the way it should be.

21:06 So, it's really nice, even [clears throat] if you're already doing this, to have a kind of like automated kind of co-pilot with you so that every time like you complete your work, you're able to make sure like we're going to check these boxes at the end to make sure we get that last mile of our work done. [snorts] >> Yeah, that's a good point. Yeah. Um, definitely the last mile, probably 10% of the work drives 90% of the impact.

21:32 So, you've heard did this thing land or um you know something similar. It's basically the that's the uh uh if you've done the hard work uh did it actually get people to be influenced by it? And so we're trying to make sure we can leverage AI to do that as effectively as possible. All right, cool. So, um let's see. So, uh, yeah. So, what we're going to be showing probably like right now is, uh, we're not going to look at any powerpoints. We're not going to do any CSS work. Um, uh, we don't need to be super precise on our prompts. Uh, and, uh, certainly we don't need design background. All of that is baked in in this, uh, in the instructions and or I guess the agents and skills in the, uh, AI analyst repo. And, uh, we'll see how that works.

22:25 Um so just a little bit of backdrop um in the repo itself we have a test data set um it's an imaginary company uh an e-commerce company think of it like Amazon um it's uh called Novarts and uh assume we've done this analysis where we saw we were looking at checkout to purchase rates for um the for the company and uh we're seeing some of these results. Now the the specifics don't really matter here again like we're trying to separate the deck building aspect versus the analysis or the analytics aspect uh at the beginning. Um what what I'm going to show what I'm going to demo next is um given these context and given these findings what does it look like when we build a deck with it?

23:16 All right so let's get into it. Uh okay hang on one second. Let me find my cloud code here. I am going to switch screen and hopefully you guys can see it. Uh did I share the right thing? Do you guys see a cloud code setup here? >> Yeah. Has like a mobile checkout conversion has collapsed. >> Yes. Okay, cool. Okay, right screen. Uh, I'm always nervous because I've shared wrong screens in the past that uh that didn't lead that may or may not have led to awkward things being shared.

23:59 Uh, okay, cool. So, um, let's see. So, for those of you who are familiar, uh, this is VS Code and uh, uh, pairing with terminal or pairing with cloud code and terminal. So for people who are not aware, VS code is a uh like a um like a work like ID workflow where you can browse your folders and files on the left um and you can have terminals at the bottom and file explorers on top whatever you want to uh make your you know almost like a desktop uh equivalent to be tailored towards your workflow. Uh and free to download. Feel free to do that. Um you you can use any IDEs to run cloud code. Um and uh you know like certainly it's uh there's anti-gravity, there's a cursor, you might have heard of all those things. Um but effectively it's just trying to make your desktop or uh makes make your workflow as clean as possible.

24:56 So uh for folks who are familiar um this is quad code in terminal and uh I am in the AI analyst repo and uh uh just you know just [snorts] showing very quickly I've got a bunch of agents skills. So agents are things like uh uh files to u to make claw code act certain way. Um they have access to different tools. they can accomplish certain goals. And uh just you know as a quick example data explorer means uh this file when you ask cloud code to act like an agent that can that does exploring the data it will be able to follow instructions for how to do that. And then we have a bunch of skills as as well. Um skills are recipes for how to do certain things. So uh things like oh what does metrics mean?

25:50 uh how to uh uh whatever it's uh uh how to do forecasting, how to explore the data, that kind of stuff. So feel free to explore those. We're not going to get into the detail on that. But what I'm going to do is so we already have the analysis done with the Novart U checkout sort of like conversion um uh conversion uh story. So, what we're going to do is um uh so I'm just going to do Hey, can you show me an ASKI diagram of the deck building pipeline?

26:28 So, what I'm asking for is show me a step-by-step diagram of how you actually do um deck building in this uh specifically in this uh in this uh agentic system here. So, it's going to take a minute to think about it. I'm on cloud uh Opus 4.7 and let me zoom in a little bit. So, it is pretty much telling me behind the scene when I use it to build decks what exactly it is that uh that it's doing. So, you can see okay here's the deck building pipeline and let me give you a visual of that. So, assuming again like the analysis phase has been completed. So it's like what's complete before deck building but once you've done um and you know this uh this AI analyst is also able to do you know framing design hypothesis explore analyze root cause all these things to ex to uh to understand what the data or the piece of analysis that you're trying to do. Uh so all those are kind of like done feeding in and then this is what it looks like once we have the findings and all the opportunities and you know like all the uh all the detail from the analysis. So you can see that oh okay uh there is a story architect that builds the storyboard. It has a narrative generator and a reviews sort of like do these things make sense and then it's got the loop of like oh if it needs fixes then let's let's go back to the architect and keep doing it until it passes these different reviewers and stuff and uh once it's done on the upper tile there then you go into you know chart maker so uh making building the actual charts to uh to really get at what the story is uh it's got like a critique for visual design. So just make sure that uh it follows the right uh the right uh uh formulas for how to how to make those look good and uh just keep doing the you know like hey do I approve it not I as a as the human but like does the system uh approve it and if it doesn't then give a feedback and keep going and then storytelling write write out the narrative and then after all those is done then it goes into kind of like putting the deck together and then so on and so forth close the loop blah blah blah so rough Roughly speaking, there's like a multi-step thing happening here where it's doing its own kind of like uh uh understanding if uh things pass the bar and then it it uh it it fixes itself or uh it passes kind of like the uh the task from one to the other until the whole sequence is uh is accomplished.

29:09 >> Maybe just to jump in there. Hi. Um, [snorts] yeah, that I think this is one of like the uh really strong differentiators is like being able to set these loops up where you work with claw to set like here's like my you know my passing criteria. you can have many different passing criteria to where like I know this deck is ready because as you if you're thinking about it when you build a deck like you you you review it yourself many times you might share it with a colleague your manager um some stakeholders that you trust and you're working all of those share out loops into your first pass right here. So, your first pass on the deck is actually ends up being what you would get like four or five passes, but it happens really quickly. Um, and it can and you can have it, you know, break and ask you like, oh, should I go through another pass or it can just loop on autonomously uh on its own. So, I think that's like one of the most powerful things uh I've seen so far when I build decks with this versus like, you know, using something else like uh Gamma, which is like an AI slide deck creator, or of course like doing it doing it on my own or doing something with like the beautify slides and and Google slides. Um they can those those criteria can all be like custom to to what your beliefs are should the the passing criteria should be. Um, [snorts and clears throat] also had a question here in the chat for you. Hi. Would the output quality be different if we use the same workflow in co-work instead of uh cloud code in VS code?

30:52 >> Uh, so the or I guess uh different uh we also have a AI analyst plugin for co-work. So it's almost like a um a companion version for co-work. um the AI analyst if you port it directly over to co-work instructions because of the differences in you know like the files and systems and the system prompts and stuff like that. Um there's also the probabilistic nature of the LLMs. So even if you prompt it twice in in the same uh in in cloud code, it may not actually produce the exact same thing. Now the data as long as it's generated from code uh will be referenced the same way the actual kind of content and how it you know phrases stuff how it frames things how it picks what's the most important thing to support a given story line could be a little different. Hopefully that's helpful.

31:56 All right, cool. So I'll move on. Let's see. Okay, so now we've we've seen what what this what this thing looks like. Uh oh, and by the way, I don't know why I have this uh chart up here. Um just to show that some analysis was was done on the mobile checkout screen. So something has happened. Okay, so uh let's see. So I am interested in trying to see what kind of themes do you have for presentations because like I want to I want to know um you know what it has access to in terms of the look and feel for the final deck itself. So it's going to think a little bit look into some of me some of the uh skills look into the directory for uh what is currently available and then give me an answer. So, uh, right now there's six on here. Um, but it's, uh, you know, like anybody can add as many as they want. Uh, it's got some of these corporate, minimal, New York Times, economist, uh, analytics specific, uh, themes and things like that. So, uh, so you know, like it's got a corpus of, uh, pretty interesting stuff. So, let's let's take a look at the New York Times.

33:09 Can you make New York Times themed presentation for the mobile checkout conversion. Let's see what story uh the audience is executives at the company. So I just gave it this prompt. Um so again like this the mobile checkout conversion story refers to an analysis that's already been done. So something like this up top here. So what it's doing is uh it's going to go refer back to the actual analysis uh which lives somewhere in under working somewhere here. Uh latest yeah it's uh outputs.

33:59 Yeah, it's got like a like a full analysis uh done somewhere here. I've done a lot of stuff here. So not going to pollute the not going to not going to make it too messy. But that's the idea. So what it's doing is uh looking back at the uh analysis and then it's going to build the end to end uh reusable and New York Times themed CSS putting together the supporting charts making the deck and then rendering it in PDF and let's see existing checkout artifact. So, it's it's found the analysis and then um and then it's going to do the actual rebuilding of the different charts and stuff like that.

34:45 And what it's doing here is it's invoking some of the skills that uh that has instructions for what a good chart looks like, how to build those charts, what the title should look like, um so on so forth. So there's a lot of best practices incorporated in some of these skills that's already explicitly been baked into um to uh to these instructions for for cloud code to follow. So we don't have to, you know, like so we don't get that AI slop that we that we saw at the beginning.

35:17 Uh okay. So it's looking at the story a little bit. Uh and then uh it's going to read the real analysis. I don't know what what uh why it was not real before, but and then it's giving me a plan for what it's doing. So, the to-do list, if you will. Um, now it's like, okay, I'm going to build the CSS, the New York Times themes. We didn't ask for any, we didn't give it any instructions for any of that. And it's going to generate the the charts, and then it will write out the deck and uh render the chart in PDF. Um, but we'll ask it for an h HTML that I can uh that I can share with you all as well.

35:58 Hopefully this goes quick. I'm hoping uh while it builds. Any questions here? I know there's a bunch of stuff that I kind of can't keep up. >> Oh, there's a question. I can kind of [clears throat] answer this one too around it says like with this workflow does claude develop predefined design templates for each theme or create fresh design layout based on the content. So like the themes we've uh predefined here if you just uh kind of have claude code uh create its own theme each time on its own. I found they're they can be pretty bad. Um so we have a few different themes that we leverage. It's pretty good at uh building themes with you. If you have something you're inspired by, like say like, "Hey, uh give me the the qualities of like an article from say like The Economist or the New York Times. I want to generate a theme that's kind of based on that um format." And then within that theme, it'll create a few different types of slides like, you know, title slides or big number slides or chart slides or bullet slides, but it still will reason through and create new types of slides on its own. So, it's not necessarily limited to that. Although, you could make it limited to that if if you wanted to. [snorts] >> Cool. So, it's run into some errors and stuff and that's all good. It's uh self-healing. So, agents are autonomous in fixing its own errors and figuring out what's wrong, all that kind of stuff. Um, so you know, like for those who are new to Claude code, um like it like pretty much until it hits a point where it can't solve certain things, uh it won't really bother you. So, it's a very compelling >> also. Uh There's one question by I think Conwell about uh the course details on the Saturday workshop that we'll have if uh there's a question about do we need it says that they could bring their own data set and how do they work with it.

38:13 >> [clears throat] >> on the Saturday workshop. Yeah. So for the Saturday workshop and also for the boot camp and also our fiveweek course, we do have a uh a practice data set we use that's based on a uh fictional e-commerce company. So you know you can think of it as like um SKUs and orders and usage kind of stuff. It's basically like a fake Amazon. Um so we'll have that data set we can use and then yeah if you want to connect your own data or if you want to work with a uh public data set that's an option as well we can chat with you in terms of like how to connect to that but um I think your question here was like if you don't have your own data set can you still do the workshop and the answer is yes we'll all start with that practice data set.

39:06 Yeah, totally. All right. So, it's still building. Uh hopefully it should be done soon. Um what we can take a look at, for example, is it generated some of the charts in the New York Times theme. Oh, now it's start to Okay, it's almost done with the deck, but we can take a look, for example, uh well, too tiny. So, uh this is one of the charts that it built with the New York Times theme. Um, you know, if you've followed New York Times, this is roughly similar to what you would have seen in the uh some of their articles. When a chart is uh embedded, it's got the different uh uh I don't know, the signature sort of red and gray kind of a kind of a look and feel. Um, and with the other skills in terms of how to tell a story and stuff like that, it actually highlights the uh like the takeaway at the title instead of, you know, whatever the description of the chart is, for example.

40:07 So, okay, cool. So, it looks like it's done. Um, so it's so it did a bunch of stuff, but basically it outputs a PDF file, outputs a mock file, used the New York Times CSS, so on and so forth. Now, I wanted to just uh Can you render this? I can show you the PDF as well. Um, but it's probably easier to do the HTML. Can you render this in HTML? And then I'll share a screen for of uh what that looks like.

40:40 This one should be pretty quick. Um, all right. The HTML rendering is pretty cool because you [clears throat] can also have it create speaker notes for you. I don't know if this one's automatically creating that, but the speaker notes is cool because you can have it uh you can set up different voices for yourself based on your audience or or who you are and uh have your kind of speaker notes um you know uh reflect whatever kind of voice guide you have set up.

41:16 >> Also uh there's a question uh from Jamie around what is the advantage of creating deck in cloud code versus cloud UI. this is uh too technical to create a deck. So I just wanted to say that I've been a uh I've not been a terminal person all my uh career I was trying to stay away from it just use directly VS code and all but as you get used to cloud code it's a very straightforward uh approach it's almost like you're chatting with it yes it gives a lot more information probably like the bash commands but uh we don't need to like understand them or uh what they mean and stuff it's almost like you're uh you know cloud UI probably uh it's just getting to understand to work with it the one thing I can say how is it beneficial than cloud UI because I have access to UI cloud UI go work and also cloud code cloud code is a lot more flexible and it remembers things so much more easily you can work with context you can go work with memory there's so many more tools you could work with it's just the starting friction I would say probably 2 three days of friction and That's it. And after that you would understand how much more powerful this is.

42:30 >> I think the other [clears throat] thing with the with you know setting up with cloud code is that you can automatically trigger creation of decks based on other things that you have going on. So if you have you know something going on in Slack or in email. Um it's like for instance like the deck here is its own pipeline but that's actually just like one small piece of our much broader more longer pipeline of our like analysis pipeline. So for us it's really nice because we can not only just it's not like I'm going to create a deck right now. I'm going to go into cloud code and create a deck. It's oh I got a question from a stakeholder in Slack. I'm going to go in cloud code and start like framing this reframing this question doing all the analysis for this question and at the very end of that it's going to automatically kick off this uh deck creation for me. So it's also just like it's just like being able to automate things um based on like other triggers is one of the kind of like key differentiators of doing something in like cloud code or codeex or or whatever.

43:35 >> Yeah, totally. [snorts] All right, cool. So, let's take a look at the thing here. Uh, so this is the deck it came up with. Um, so the uh it's got kind of like the um bit of a look and feel for the New York Times things, although it should should have been a little wider, I would say. Um, but anyway, uh, so the deck is pretty much kind of like going through the tailor for an executive audience. It is done the minimal almost like the minimalist uh way of uh showing the information that is relevant for the audience itself. Notice there's not a lot of clutter here for example uh and uh what were the findings so on and so forth supporting ticket all of these were done by cloud code like I didn't touch I didn't give it anything in terms of um in terms of the the kind of like the display requirements and things like that and so um you know like uh and then it goes to what we're asking for for an executive audience typically it's like okay uh we're looking to ask for a certain uh certain things and not just you know tell a story and leave without uh without any sort of next steps. Um so hopefully this is uh helpful and you know it's got the appendix. Um but this is a pretty short deck that it created because the audience itself is uh uh is uh is pretty tight.

45:06 So that's kind of like the that's kind of like the idea here. And uh we can obviously you know like if we switch back to the switch back to cloud code we can obviously if you remember just uh 10 minutes ago or so we have the ability to swap themes right like we have the uh economist theme for example or the corporate theme um and so it would just be a matter of like you know hey I don't like this theme let me change let me swap that one prompt to clot code in the same instance in the same session and and then it will it'll do the rest and then you go check and you go iterate and you you uh give a feedback and just keep doing that. So that's kind of like the uh the the the idea here.

45:53 All right, let me go back to here and I know we're coming up on time so uh Okay, cool. So we did the demo here. Um, okay. Before we get into Q&A, uh, so couple ways to, you know, like we have pretty limited time in, uh, in the lesson, uh, but couple ways to continue to take the next step. If you're interested, um, this Saturday, so two days from now, we have the intro to claude code analytics uh, workshop, which is three hours for Pacific time people in the morning. hands-on workshop. We'll help you install cloud code, clone the repo, run real analysis, and uh we'll basically get you set up.

46:35 Uh we're doing it almost for free here. Um uh we're doing kind of like a $25 with this code uh because we do want people to show up and not just, you know, like sign up for free and uh and and and not uh not take it seriously. And then we have in another week we'll have a two-day boot camp um cloud code analytics. So where we help uh where we work with you to actually build a system so you can actually build something like an AI analyst yourself that's tailored towards your own use cases, your own context and uh and your own expertise.

47:10 So we'll show you and walk you through how to do that. We've done a cohort of this and uh we've got really great reviews uh off of it. So, uh, we'll send over the link in the chats, uh, as well, but, uh, feel free to kind of like ping us, uh, for any of these if you have questions. We have a promo code here for 20% off of that of that boot camp that happens in I guess a week and a half now. Uh, expires Saturday Pacific time.

47:38 Uh, end of day Saturday Pacific time. So, uh, definitely take advantage of it if you're, uh, if you're interested. All right, we've got 10 minutes left roughly. So, I think we can probably [clears throat] open it up for questions. Can read through some of the ones on here. I think Nikki just had a question. Nikki, do you want to ask your question actually? >> Oh, it's a really good question. Yeah.

48:09 >> Um, so I was curious um how does um this workflow differ from using claw design? um particularly because like right now I have some free credits that I can try things out. It's about 20 prompts worth. So I really want to get value out of that. And then also I want to keep the cost slow with whatever workflow I come up with. So I'm realizing claw design can't be like that one place that I stick to.

48:35 >> Yeah. Um I tried cloud design pretty extensively myself. Um there's certainly a new product and I'm I'm guessing you're talking about the slide deck builder component of cloud design, right? >> Yeah, exactly. But we are hoping to like uh build out our design system and start using that to rebrand old materials and create new ones. And I'm starting to realize like oo that's going to get expensive. So I want to come up with like the most efficient workflow I can but also take the val you know take leverage the value of this new tool.

49:08 Yeah, totally. Um, yeah, Cloud Design for those who haven't used it, it's uh the new um the new system that Claude rolled out probably like two and a half weeks ago now or three weeks ago. Effectively, you can go to claude.ai/design, something like that, where you can build interactive apps or do anything related to design. And building slide deck is one of those. Um, I built two decks with it and uh I exhausted all my all my credits and I think over time it's gonna get really good definitely. Uh that's that's no that's no question. Right now it's pretty painful to do. Um so uh instruction following uh understanding your intent understanding your context that's been pretty rough. Um I would say now over time that would be different.

49:59 uh the you know like what Sean got into earlier is that in code it in clot code specifically in terminal it's just a lot more flexible and uh and you know like I think the cost is a lot more predictable in in in a way uh and so you know like pretty sure design cloud design is going to keep getting better and it's going to keep getting uh like being able to follow instructions and all that kind of stuff uh improvements pretty well over time. Right now it's pretty hard to do. Um and you know again like you lose the ability to sort of like chain different workflows together.

50:41 Um uh you have a very singlepurpose kind of like a a workflow uh when you look at something like a cloud design for example versus you know cloud code and terminal. you can chain it with whatever you're doing all together and just perhaps like oh now build me a deck after some of this other stuff that you that you've done. I don't know if you guys have anything to add. >> Not for me. [snorts] >> I think there are a couple of questions around uh it looks like I don't know if you've covered it in the previous one.

51:17 It looks like AI agents can operate here without being invoked by slash commands in the terminal. Have you built MD files in the background to be able to respond to you? That's one question. I think another question this was by Anish. Uh uh and there's another question around something similar is about how did you get uh claude code to get the analysis for you done? like did you like you know share how how is the analysis fed to plot code in which format was analysis or data uh is stored so I think both are kind of similar so just want to do >> yeah Sean do you want to answer the slash command one and I'll take that second >> yeah the question is like had have we built agents [clears throat] and slashcomands in in the repo to do stuff yeah we have like so this repo is like uh it's like dozens of uh agents skills slashcomands. Um, and so a lot of those will trigger automatically based off of like phrases that we've put in there where you'll it'll look if there's like, you know, semantically similar phrases like I want to create a deck, I want to create a presentation. um it'll automatically know to kick off that skill which will then um invoke these these agents and this entire workflow.

52:39 You can also trigger them uh directly. Like sometimes, honestly, a lot of times when I use uh any sort of a genetic system with cloud code right now, I'm so paranoid that it's going to go off the rails that I'm like I first prime it to say like, "Hey, tell me and and how kind of did this, right? I was like, tell me about the uh pipeline we have in this repo that um that can build out a deck for us." and then okay go go go leverage that pipeline to do it because sometimes it will you know escape the harness as well. Um but yeah you can take a look at the whole kind of uh open source repo to see what's in there and then we're going to obviously in our our boot camp [clears throat] in a week and a half uh we go into how you build all of that from scratch for your use case. what our kind of methodology was when we built this for ourselves.

53:39 Then we also have a fiveweek course where we spend the last week of that fiveweek course purely on um presentation and we kind of talk about not just the uh the agents and skills building but the kind of like frameworks and stuff um behind them. [snorts] I don't know if that answered your question. And then the second part of the question was how does Claude know where to pick up the analysis? So the analysis was pre-done. Um again the AI analyst itself is a full endtoend agentic system that goes from the very beginning of um you know hey given a question how should we reframe it? How do we what do we look for and uh what is actually useful? That kind of stuff all the way down to hey here's the finished deck. And so um it has all the different wirings and uh again very similar to what Sean has said about uh automatically triggering invoking um different commands and skills. Uh same thing like if it's like oh like uh hey I've done this analysis and then I'm referring to that analysis then it knows kind of like oh okay like here's what I'm looking at uh and let me let me see what what that looks like and then you know it passes over and like continues the with the pipeline.

54:58 Hopefully that makes sense. Um, another question for the Saturday workshop. Are there any prerexs such as software tools that need to be installed beforehand to get started? So, this is that intro to cloud code analytics. Um, it's a threehour workshop we're doing this Saturday, May 16th. This is like um like I think we said this in a couple things in the chat like we like we find that like anyone can really uh leverage cloud code um for any sort of use case that they have like just no matter what their technical ability is but there is some kind of initial friction in getting things uh set up. So that's kind of what this whole Saturday thing is about. It's like getting folks set up hands-on with live instructors and kind of like unblocking people so they can run through uh like the analytics repo or or start thinking about what they want to do on their own.

55:59 Um so that being said, there's no prerex. Our expectation is that you you could have never used any of this before. Uh you'll need a computer, obviously. You'll need a you'll need Wi-Fi to be able to access Zoom and also to hit the anthropic API and then you'll need a Cloud Code Pro subscription which is uh $20 a month currently. >> Best if you have a Mac. Uh we are we don't know PC much. Um but we've had students who worked work themselves out before as well. Yeah, we'll spend some times on how you can we can translate the code to uh from Mac to whatever PC or or Linux or whatever.

56:48 Um, another question here or Emma, do you want to read your question actually? Yeah, my question was just sort of around the repo that you have and how I can take pieces of it and incorporate it with the repo that I'm already working with at my company. But I saw Nikki's response and can probably just ask Claude how to do it for me. [laughter] >> Yeah, I think it's pretty cool how how portable things are. So, um, yeah, at my work, we all work on like separate kind of repos locally and then kind of have like a master one and we're not kind of pushing to those as we typically would with like inkit because it just gets things get so messy and and different.

57:39 So, what I usually do is like I'll have cloud code be like, "Hey, go look at that repo. Look at mine. Tell me everything that mine does better than that and and it does better than mine. Tell me all the the gaps to fill and now create a plan to basically like bring mine up to parody or to add any to fill any of those gaps. Then you can review the plan yourself and have it execute it.

58:06 >> Okay, awesome. Yeah, that's super helpful. Thank you. >> No problem. [snorts] Good question. I saw Max has a question about the token or not token the how if if the $20 plan is uh is good enough. Yeah. So short answer oh Shabby already answered yes should be should be should be more than enough. Um >> price average up. Yeah. >> Yeah. Especially Claude uh I don't know if you guys have seen Anthropic just announced that uh uh pricing war with uh Codeex. So they actually upped the limit for both the five hour window and also the uh weekly limit by 50% or something like that or double or something like that. So um it it actually is quite abundant uh if uh if you kind of think about it that way versus before which uh uh is easily running running out. Um, so far I think I think now is probably the best time to kind of learn because you can burn not, you know, like not not wastefully, but like you can burn compute in a way that you can really get into the learning without shelling out for the max plan or or or things like that.

59:22 >> Um, see, we got another one here. Uh, pipeline slide from before the numbers weren't vertically aligned. Is that limitations in MARP or wondering how good the slides can get? Pipeline slide from before numbers were not vertically aligned. Oh, I see. Uh, yeah. So, the fifth step if you remember the iterate piece. So, you can actually give Claude the same feedback like, you know, hey, I see that you aren't doing this or like the thing is out of whack or like uh text text boxes were spilling over the page and things like that. Uh it it actually knows how to fix itself or it knows how to review it and then fix itself. So, it is by no means a limitation on any of the any of this stuff. Um, as long as you sort of work with it like a person would, uh, it will figure out kind of like the stuff that you're referring to and that it will do the uh the the the the fixes for Yeah. And then and then you just want to make sure that uh whenever you do those fixes, you want to codify them. So, uh, because it'll default to like oneoff fixing it for that deck you're working on after you get it to a place you want it to be. Uh, usually like kind of my next step is like working with CL like, okay, what was the difference between what you were doing in this this final step versus that, you know, draft before that? how do we update uh the existing kind of templates or uh uh skills or agents or or whatever so that the next time those are all aligned from the first place. And like every time you build a deck, there'll probably be some little tweak like that that you have. Um but it's really cool because the next time you use it, it just gets better every every single time.

61:21 >> Yep. Uh, I know we're out of time, but I can stay for a couple more minutes if people have more questions. >> Um, Sheila on like are folks using what laptop are people using for cloud code? Is there recommend to get separate equipment? Um, so I mean like cloud code itself like everything that's running is running on anthropic servers. It's not running on your machine. So like your computer doesn't matter that that much in that case. But like if you start if it's starting to like create a bunch of output um on your machine, if it's starting to like call other like you know for us like we do a lot of like analysis works as calling different stats packages and stuff or if you're starting to run a lot of things in parallel that's when you want to kind of like um think about graduating to a different machine. Um, I don't know. I we me highest Ravi I'll run max. Um, but I would say like, you know, I don't think you need to go shell out and buy a new computer right away. Like see what you can get away with right now and then yeah, it might be worth buying something.

62:48 >> Very cool. Thank you. I [laughter] after a layoff, I'm at home. I've got my old Dell. I'm like, "Oh boy, blow the dust off of this thing." blow the dust off, try to remember the password. >> Yeah, I'm I'm rolling in it. But, um, you know, even for security purposes, right? Like, I'm just getting to the point where I want to start toying around with automating some of my personal stuff and I'll, you know, start with some, you know, job search or business ideas. But now I'm going, "All right, I know I I need to do like make sure I'm double down on my password stuff, but does it make sense to have a separate machine anyway?" From what I'm reading, it it sounds like it. And the Mac Mini is popular, but I don't know.

63:34 It's a little It's a little iffy out there on the advice. >> Yeah, really depends on what you're doing with it. I think uh you know, the Mac Mini Crew is uh Mag Mini is really powerful, by the way. Um I I've been playing around with it for the last week and a half for you know high really high bang for the buck. >> Um the whole you know separate computer thing I think pertains more to for example like if you run something like an open claw or the Hermes agent where uh you know like it's doing things directly itself uh without you being in the loop which could be pretty dangerous if you think about it. So um that's probably the context that you hear it from. Um, yeah, you know, for the big biggest bang of the for the buck, it's a Mac Mini is great regardless.

64:20 >> Okay, cool. I'll take a look into it. Thank you. Yeah, was it looking at Open Claw, but we're just going to slow our roll and just take it step by step. [laughter] >> Yeah, Open Claw in the grand scheme of things is like level five versus let's say like a cloud code is probably like a level three in terms of complexity. >> Awesome. Thank you. >> Cool. Okay. Uh, is there anything else? Any other questions?

64:55 Um, student asked about the time for intro to cloud code analytics on Saturday. That's 7 a.m. to 10:00 a.m. Pacific time. So, Saturday morning, West Coast. Um, >> one last thing on the workshop. So, I have um an engagement that would cut me off about halfway through. Do you have these workshops frequently or if I did it and I was there for the first hour, hour and a half and then is the rest recorded where I could just kind of catch up on it?

65:28 >> Yeah, the rest will be recorded. And I think that first hour, hour and a half is probably a good one to be there for because that's when we're going to kind of like [clears throat] demo the capability a bit and then like set people up. Um, and then you can follow along on your own leisure. We haven't ran so we have a bunch of we have a couple two-day boot camps that we run monthly. Uh, we haven't ran this intro to cloud code analytics one before. This is the first time we're running it. Um but if uh yeah if it goes well I'm sure we'll run it again but either way you'll get the recording.

66:02 >> Yeah. >> Okay. >> We we had we generally want people to come in person and participate live. Most of the people did but there were few people who did async for the boot camp as well and they they still um like you know like the work and were able to get a lot of value but yeah >> great thank you. Yeah, I'll see if I could move my other engagement a little bit, but >> sweet. Cool.

66:33 Okay, if uh >> if nothing else, thank you everybody for joining us. >> Yeah, we'll see we'll see y'all in the Saturday workshop for those who are going to make it and then hopefully see you some of you in the uh boot camp in a week and a half. If you got any questions, feel free to reach out to us on LinkedIn or or email us or go to AI analystlab.ai if you have any questions. Happy to um or get hop on a call with you as well if you have any questions around the boot camps or workshops or anything. Then we have our our Slack community too.

67:10 >> Yeah. Oh, I forgot completely forgot about the Slack community. >> Yeah, I I shared it. I shared it. We can probably share it in the email as well, right? When we share Yeah, we'll send it we'll send an email out after this with all the kind of details to that stuff. Um, but the Slack community is a nice way to DM each other so much easier. Yeah, cool. All right, see you guys around. >> Thank you everyone.

67:31 >> Thanks for staying. See you. Thanks guys.