# 165 I BLEW UP Instagram With AI Content (4M Views in 30 Days)

Show Notes

I sat down with Nick Puru to break down the exact Claude Code setup he used to pull 4 million views and ~6,000 newsletter signups in the last 30-45 days for his AI consultancy. Nick walks through his short-form content factory inside Claude Code — the CLAUDE.md "brain," his ICP file (an avatar he calls Patrick), his foundations doc full of algorithm lessons like "negativity always wins," and the three skills he uses to write captions, generate scripts, and review them. We get into how he tests three hooks per video with Instagram trial reels, the humanizer skill from Bader on GitHub that strips out AI tells like em-dashes and "it's not X, it's Y," and why he treats every Claude project like onboarding a new employee. By the end of this episode, you'll have a clear blueprint for building your own short-form content system in Claude Code — starting from a single CLAUDE.md file and expanding from there.

Timestamps
00:00 – Intro
00:37 – 4 million views in 30 days
01:53 – Reels to ManyChat to newsletter funnel
03:31 – Tour of the Claude Code folder structure
05:02 – How the write script command works
06:29 – Why human-in-the-loop matters for content
07:38 – Inside the CLAUDE.md brain file
08:11 – Meet Patrick, Nick's ICP avatar
10:04 – Foundations file and "negativity always wins"
13:43 – Live generating a Claude Cowork script
15:44 – Testing three hooks with Instagram trial reels
18:48 – Most bare-bones version to start with
22:36 – Why ICP and brand voice are foundational
24:28 – The humanizer skill from Bader on GitHub
27:01 – Treating AI like a new employee
29:35 – Context beats prompting in 2026
31:13 – Where to find Nick

Key Points

Nick's short-form system drove ~4M views in 30 days and ~6,000 newsletter signups in 45 days for his AI consultancy — every video CTA pushes a lead magnet via a ManyChat flow that collects emails and hands off to an appointment setter.

The whole system lives inside Claude Code as three skills: one writes scripts, one reviews them against a quality checklist, and one generates captions. Trigger them by typing things like "write script" in the terminal.

The CLAUDE.md file is the brain. It holds the output format, writing rules, key principles, and a pointer to an ICP file built around an avatar named Patrick — a small-to-mid market business owner ($few hundred K to $15M, 2-50 staff) who has tried ChatGPT once or twice but doesn't know Claude Code.

One of Nick's foundations is "negativity always wins" — the algorithm rewards a stronger emotional charge, but he warns against using it on every video or audiences pick up on it. He calls it in only when the angle fits.

He tests three different hooks on the same body and CTA using Instagram trial reels, treats it like A/B testing titles and thumbnails on YouTube, and feeds the winners back into the system as analytics context.

To kill AI tells like em-dashes, bullet points, and "it's not X, it's Y," Nick runs scripts through Bader's humanizer skill from GitHub. Corey adds a similar instruction in his agents.md to never use dashes.

Treat Claude like a new employee, not a magic box. You wouldn't expect a hire to crush it on day one — you'd give them SOPs, business context, your website, and an ICP. Same playbook for building any Claude project, whether it's short-form, long-form, or LinkedIn posts.

Bader's humanizer skill on GitHub - https://github.com/bader-research

ManyChat for Instagram DM automation - https://manychat.com

FIND ME ON SOCIAL
X/Twitter: https://x.com/coreyganim
Instagram: https://www.instagram.com/coreyganim/
LinkedIn: https://www.linkedin.com/in/coreyganim/
YouTube: https://www.youtube.com/@coreyganim

FIND NICK ON SOCIAL
YouTube: https://www.youtube.com/@nickpuru
X: https://x.com/nickpuru
LinkedIn: https://www.linkedin.com/in/nicholas-puruczky-113818198/

What is Build With AI?

Most AI podcasts talk about what's possible. Build With AI shows you how it's done, live. Each episode, host Corey Ganim brings on entrepreneurs and operators who share their screen and build real AI automations, workflows, and tool setups right in front of you. No boring slides. Nothing that hasn't been battle-tested. You'll watch actual implementations get built from scratch so you can follow along and do the same in your business. If you're a non-technical entrepreneur who wants to put AI to work without becoming a developer, hit play and build along with us.

Corey Ganim: All right, Nick, what are we building today?

Nick Puru: Go on everyone. We are going to be showcasing my short form cloud code instance. So in simple terms, this just helps me create short form content. So I start with an idea. I see somebody else's video. It inspires me. Or if I just have an idea myself, I run it through this AI system, this AI system, however you want to frame it, and it's going to create content for me. So I could actually go ahead and already share my screen. and just showcase some of the results that this has been getting thus far for my Instagram.

Corey Ganim: Yeah, let's check it out. know short form is something that a lot of people either they just put out bad short form content or they don't have a system in place. So I'm excited to see how you do it. And obviously you've been crushing it.

Nick Puru: Yeah, absolutely. So some quick background, like everything that you're seeing on the screen right here. So obviously like the big things that stand out, like the amount of views. So in the past 30 days, we're able to amass about 4 million views. ⁓ Scroll down a little bit further, all the interactions, but the big standout 4 million views. So with that being said, I mean, there's a lot of, a lot of numbers that we can crunch that this is attributing to more than just views. So being able to, so actually just to give you a little bit of background about what we do, some context about us, like we run a AI consultancy. we drive pretty much all these views into our newsletter. So I think it's about close to 6000 signups that we have acquired within our newsletter just within the past about 45 days or so, where now we're nurturing everybody who's watching our content, then flowing to our newsletter, eventually, ideally, they become clients of ours. So very high level, that is what we've been able to amass those far. So with that being said, be able to jump into what it actually looks like inside of Claude code and how I actually use this every single day.

Corey Ganim: And now when you say you're converting them to the newsletter, are you just doing like a CTA at the end of the reel that says like, comment XYZ and I'll send you just like a lead magnet through mini chat? Is that kind how you guys are doing it?

Nick Puru: Yeah, exactly. So if you could still see my screen here, you can see if I just click on one of these, for example. Now, I'm not going to put the audio on, but at the end of this, I'm usually going to say something where it's going to lead them to, all right, for example, this one in particular, it's a guide on cloud cowork. So I'm saying, Hey, if you want to, if you want access to this guide, then just comment, cowork, and I'll send it over to you. So sometimes I'll do something that's going to be very specific to what the content's going to be. Other times I'll just be saying like, if you want access to my weekly AI newsletter, then comment this and I'll send it over to you. But anyways, we have a many chat flow set up where it's going to be essentially just discussing with them, collecting their email and sending off a few other things inside of DMs. And then I'll have an appointment setter take over from there once we're asking an open ended question. ⁓ But all of this, it is stemming from the video.

Corey Ganim: Got it, okay.

Nick Puru: as the trigger, I guess. ⁓ Cool, so, let me go ahead and show you what it actually looks like. Let me figure out how to share my screen here. If I could share my entire screen, would imagine I would have to. Let's see. Okay. So here's what it actually looks like. I pretty much just live inside of the terminal whenever I'm using this. I don't deploy it on Telegram. I don't really play it on anywhere else. It's very possible to put this as you already know. You can put this on WhatsApp, Slack, iMessage now, really wherever. But anyways, I just use it inside of the inside of cursor as close as the terminal so you can get almost. But anyways, here on the left hand side, if you're not familiar with cursor for anybody who's watching or or Claude code or any IDE. This is essentially just my folders. So we have all this broken down. We have our Claude file. This is all of my skills. So I'll try to give a very high level rundown of this first and then we'll dive into each one and elaborate on them a little bit further of all the files. So Claude, this is just consisting of some of the different skills. So I have three specifically. So one, that's creating captions for all my videos. Two, reviewing any script. three that is writing all the scripts. So whenever I want to create a new script, whether it's an idea, so I could start off saying like, Hey, I have an idea about something, whenever it may be. And then eventually let's say you want to turn that into a script. All that I have to do is just write script and it's going to start creating it for me. So pretty straightforward on how to actually start using it, but it is very, uh, it's a very meticulous setup that I actually have in place. And this is something

Corey Ganim: So you could just say like, for example, you know, backslash write script about Claude cowork and it'll just go and research Claude cowork and then write a script. I'm assuming based on. Formats that have been successful for you already, right? Like you're giving it references that a videos that have performed well.

Nick Puru: Yeah. So what I typically do is yes, I do provide it with some, some previous analytics where here's what works well. Here's what doesn't perform well, things like that. But anything that's going to be a little bit more specific about like maybe it's Claude Cowork or anything else, like any, anything that's more nuanced, I do have it do a little bit more research. I give it some more access to my recent videos saying like, Hey, I just uploaded a few videos on this topic. Here's what did well, here's what didn't do well. So let's try to take what we learned from all of this and create a better version. ⁓ That's more so how I have been using it as of late. So it is kind of a dance where you do have to, know, AI is doing about 90 % of the work. Don't get me wrong, but there is a bit that you have to configure and tweak and make sure like you are providing the proper oversight. I think that's kind of where we're at generally, even outside of ⁓ using systems like this. using AI like practically in any setting. It's really just having some some solid oversight and making sure that you're steering it in the right directions, right?

Corey Ganim: Right. Yeah, there's, there's going to be a human in the loop, especially when it comes to content. feel like if, if you want to put out good content, there has to be a human in the loop. Like you can't just have the AI go wild.

Nick Puru: No. Yeah, absolutely. ⁓ So it's the same exact case with this. Now, moving on, we just have our directives inside of my directives. And ⁓ this is kind of, I called it directives as this was ⁓ what I used for a PCE framework that I called, where basically this was just going through ⁓ some processes that Cloud Code now optimizes for, so completely irrelevant. But anyways, the directives, this is just the some high level instructions. So of course you'll recognize we have a Claude MD now for anybody who's not familiar with Claude code or this entire environment Claude MD, this is the brain. So in the brain, I'm providing it with high level instructions. And actually as of yesterday, there was a leak where ⁓ Claude code's entire code base was released. So the big takeaway from that. is we see like how you should actually be creating your Claude environments, how Claude is actually set up properly. So with that being known, we know how we can actually play around with Claude, how we should be polishing and setting up everything. So as of just yesterday, I've just refined this to be adhering to that. So how that actually looks is I just break things up into how I want the format of each video, video scripts to be outputted. I provided some brief information on my audience, so I call him Patrick.

Corey Ganim: I saw that, yeah, I was hoping you were gonna go into detail on that, that's interesting.

Nick Puru: Yeah. So Patrick is my ICP. So Patrick, consists of small business owner, small businesses to mid market. So anywhere from a few hundred thousands to up to $15 million in revenue, about two to 50 staff. So I want it to create content that is going to be speaking to Patrick. Now it's not always that we're going to have it be so specific, like anything that's going to do just be talking to like a small business. Like we do generalize our content a little bit as opposed to just going like bottom of the funnel. ⁓ but of course, you know, we're just providing some more, some more language saying like never say Patrick in your scripts. Don't use jargon without explanation, stuff like that. I mean, I can go on for days just giving AI instructions about this stuff, but it's very important that we're keeping our Claude MD file pretty loose and Very direct and punchy. Then we have some more information. So writing rules. So what it should actually look like. ⁓ So focus on like anything that's going to be benefit, make sure we're talking about and speaking to any results, speak to Patrick specific context, not just like business owners generically. ⁓ What else beyond that? We're always trying to showcase demos first. And then we have key principles. So going through some things that I've just learned. I mean, this has been. This entire environment has been something that I've been building up for ever since I started short form content. So maybe about a year, even some, some key lessons that I've taken from, from long form. YouTube I've incorporated into here. So, I mean, I've been building this for effectively about two years with obviously not using it all inside of cloud code, but building all these lessons, all these takeaways. that is what is essentially being inputted into all of this. So that's what these trinity of values, the ⁓ quality of our content is looking like, not being the AI guy or being the AI guy for business rather. Yeah. So within this CloudMD file, we kind of give it some, we're pointing it into specific directions. So we say, okay, you can reference these specific files when you needs to. So if we close out this, we can elaborate on some of those and what that actually looks like. So for example, let's take a look at this foundations one. So within the foundations, we just have some like, I mean, some of these are very nuanced. Some of these are pretty general, pretty broad where if we just pluck out, let's look at this first one. So negativity always wins. So core principles, this is negativity. It speaks, it speaks volumes. So it is going to have a more resounding effect on somebody.

Corey Ganim: Mm-hmm.

Nick Puru: if they were going to listen to something that's going to, something that's going to, ⁓

Corey Ganim: You

Nick Puru: insult them. So for example, somebody's grandma is a fat bitch. Feel free to blur that out. But ⁓ if you're saying like somebody's grandma is a sweet old lady, it just so happens that the algorithm, it loves things like this. It just speaks to the negativity. ⁓

Corey Ganim: Right. There's just going to be a stronger emotional response, kind of like you're saying to something super negative versus even something super positive, right? It's emotional or I guess negative emotion is a stronger driver of, ⁓ that causes people to react, which is I think why you see rage bait content, you know, for better or for worse performing so well.

Nick Puru: Yeah, exactly. ⁓ So it's it's exact things like this that I'm trying to input into pieces of my content, but I don't want it to be like every single video I'm saying like, you suck at this or you're stupid, things like that. Like, it's just going to be a little bit repetitive. And people will pick up on that. So I don't want to go with that route. But it is a good factor to include in some of your content. So sometimes like if I if I just know that it's going to be good angle to take prior to actually creating a video. let's say I have an idea and it just so happens that like I thought like, okay, well maybe the negativity is going to work really well. So I'm already set on using that. In which case I could just say I want to create a video on Claude cowork and let's go with the negativity. So if I were to say something like that, it's just going to immediately find in its files in the environment within this foundations that is going to leverage this negativity always wins ⁓ premise. So in that case, I mean, we could test this out, see what's gonna happen. ⁓ Might not be perfect because usually I give it a little bit more direction and ⁓ what I actually want the video to be about, maybe like the results that I'm trying to get, maybe I'm pushing a For example, we just recently had a webinar or workshop where we were talking about how to effectively use cloud code in a business setting. So in that case, we wanted to push a lot of our CTAs, all the lead magnets to our workshop. So that's where we will be a little bit more specific about the results that we're trying to get, or maybe if we saw like another news piece that we want to speak about. we'll just give it some information, give it the website, it'll go ahead and go research that. ⁓ But again, it just comes back to giving it some proper oversight, giving it some good directions. really the premise of ⁓ using AI effectively inside and outside of things as specific as this current system. So it is running for quite some time. mean, already taking a minute. So I mean, I guess that just speaks to how much it actually has to go through within this. And I mean, sometimes that can actually be detrimental. If you, for example, if you have

Corey Ganim: Right.

Nick Puru: Claw.md if this is too long. It's going to hallucinate. It's not going to remember like what it should be focusing on, what it shouldn't be. ⁓ But anyways, we could see phase one. So it's coming up with the strategic decisions. So the hook, we're gonna start off broad. So like delegating task to AI on your computer. Let's see what else. So it's finding this negativity angle. rather than the standard tool just dropped opener that I often go with, that's something like that. It's top of funnel, those typically do very well. It's going to lead with the pain of repetitive computer tests manually. So it looks like it's trying to play this into the angle of doing this towards Patrick, so some sort of business owner. So phase two, hook options. So you're still doing manually.

Corey Ganim: Hmm.

Nick Puru: You're still manually doing tasks on your computer that AI can do now, and it's costing you hours every single week. Stop spending your mornings on emails, spreadsheets, et cetera, et So it doesn't actually look like these are too aggressive of negativities, ⁓ but you get the idea. So.

Corey Ganim: Right. But that is a good angle. The third one, there's a good angle where it talks about, cause that's a pain point that every business owner has where it says every hour you spend on admin work at your computer is an hour you're not closing deals or serving customers. Like regardless of, obviously that's not the most negative statement in the world, but it's definitely going to get the attention of a Patrick, like a small business owner who's super buried in the weeds and needs to buy back some of their time with AI. So I mean, you know, I think it's still an appropriate hook.

Nick Puru: Yeah, absolutely. And what usually happens is, so let's say I go with one of these or one of these I prefer over another. So with this either I'll keep tweaking it saying like, maybe we could change like this, this minor point or whatever it may be. And ⁓ just keep refining it from there. I often touch different hooks. So maybe I take all three of these hooks and I'll record them. ⁓ I'll record them. record the entire script, just do like three different hooks with the same body copy, same CTA, and just test like which one's working. And then from there, like I'm going to take the feedback, take the analytics that I'm getting from Instagram. There's something called trial reels. I don't think I'll be able to pull up my computer. think I'll have to have it on my phone, unfortunately. But with trial reels, you can test ⁓ what's going to get pushed to. Essentially, it's just going to see which one's going to be performing better. with YouTube, I it would just be comparable to the A-B testing of like titles or thumbnails. anyways, yeah, so that's the foundation. So I'm just going on more concepts. So the first one, we have that quick win. I mean, the list goes on and on of different foundations. So these are all just different things that I learned over time. about the algorithm, things that I have learned from different YouTube videos. So it's not always that it's following these foundations because again, it is very long. So it's not inherently a process of this to go through this foundation. It's usually something that I have to have it referenced specifically.

Corey Ganim: Now for that foundations file, because that is super long, like you said, I'm assuming you didn't just write that from scratch. Is that something where you kind of maybe just brain dumped into Claude, like everything you've learned and it just compiled it into this, this markdown file here? Or how did you go about actually building that?

Nick Puru: Yeah, I mean, more or less, that's exactly what I did where it starts off with a brain dump or it's just other files where I have some notes of different things. So for example, like I knew I had something written out two types of reinforcement. So there's positive reinforcement. There's negative reinforcement. So both are trainings to like a predictive behavior. I mean, this expands into something much larger. And, ⁓ but yeah, like this, is extremely long. So sometimes I will. better way to go about that is. I have some original files or I just have some brain dumps that I'm going to be putting into, whether I inserted it into here or just like through the native browser inside of Claude. And it's either going to expand on it a little bit, make it a little bit more. Depictable for the AI to actually understand. So to answer your question, yes, that's more or less what my products look like.

Corey Ganim: Now, what is the, so this is an awesome setup by the way. And clearly it works because you're at almost 4 million impressions in the last 30 days. So if somebody wanted to do like the most bare bones version of this as possible, just to get it up and running and so that they can start testing and tweaking and getting feedback in terms of data, where do you recommend they start? Like which of these files do you think they should create first?

Nick Puru: Yeah. So first just assess what you currently have. So whether that's some notion documents or maybe you have some SOPs, whatever it may be, just assess what you have. If you have information about your business, any context, so start putting that in a drive, put it somewhere just so you can easily access it at some point. Because what I'm about to get into, that's where you're going to be leveraging it. And that is the Claude MD file. So It all starts with this. Like this is the entire brain. This is the file that is arguably the most important. So right here, like what I walked through, it's very high level, but I mean that fundamentally is, is all this is. It's, it's a very high level file of, of what I do of how I want this to output anything, any, ⁓ any scripts, anything like that. And then from there, so once you have your cloud.md file, and this is where to actually build this out. You use AI, AI is your weapon. So use AI to build out your CloudMD file. Take a look at the Cloud leak. Analyze the lessons, the key takeaways of how to actually structure CloudMD file. And then from there, you could expound upon that. You could give it some more direction. So for example, I have a quality checklist. So number one, like is your current content consistently excellent? It's really just a list of things that I needed to run through before it actually starts, it finalizes anything. And that has to do with my, if we open up my commands, that has to do specifically with my reviewing script skill where it has to leverage the quality checklist. But anyways, getting back to the point, you just provide your CloudMD file. provide your Cloud Instance with just context about what you do. So of course, for all of, or any of you guys watching, like this may or may not be directly related to content, whatever it may be. Like you just want it to provide it like your goals. It's responsibilities, like what you want it to do, context about you. So for example, when I have to provide it, one of the most important things that I have to provide it, like who my ICP is. So again, like that's going back to... this and there's actually another folder. This expands much. It's just a more elaborate. ICP breakdown where it elaborates on Patrick. So who, who we're trying to speak to. So this is of course a very core component of creating content. We want to be speaking to like these specific industries. want to like, it has to know like they can't be too technical. So maybe they've heard of chat GPT. They tried it once or twice. Maybe they don't know too much about cloud code. So that's telling the system you can't talk too much about cloud code. So it's as simple as just starting with a cloud and MD file from there. You're inevitably going to get into a rabbit hole where you're finding different, different little tweaks, different adjustments you have to make. You're constantly going to be learning. And with this, you're just going to accidentally become more and more technical. So, ⁓ that's not to say you guys aren't technical now. don't exactly know your audience, but anyways, It all starts with just providing it context and the outputs that you're trying to get. It's as simple as that.

Corey Ganim: I think you, I think you're spot on when you talk about how important the ICP file is as like a context document. That's something I tell people the, two foundational context documents, whether you create them as skills or just, you know, it's just general files that you're using inside your AI projects. The two most foundational ones are ICP and then brand voice. So I'm sure you have something in there about your brand voice, like how you talk. things you like to say, words you avoid, right? That type of thing. So that it actually sounds like you and not just like AI, but I feel like those two pieces of context are, are two of the most critical. And then everything on top of that is more specific to your setup or what you're trying to do. Like, would you agree with that? Or is there any, anything else you think is like super foundational?

Nick Puru: Yeah, 100%. So first off. This kind of relates to something a little bit more expensive than just like my, my exact setup. So overall, yes, I do provide it with like my brand voice. I don't give it like specific words or anything like that. Generally, like I have, I give it like my style, my tone, ⁓ anything like that. So I have, I try to go with more of like a, ⁓ include some humor, some, some wit within my videos. So that's kind of like the style that I take, but yeah, I mean, that goes for, for.

Corey Ganim: Mm-hmm.

Nick Puru: legitimately anything when you're utilizing AI, especially like if you're using this for writing emails. I mean, this is a huge thing for any of our clients that we've worked with or anybody that comes to us saying AI sucks, can't write emails, for example. It can't figure out my language. And it's like, he even tries to instruct it properly. So yes, you do have to provide it with... What that actually looks like is just providing with examples of...

Corey Ganim: Right.

Nick Puru: You know, your tone, some past emails, for example, you can provide it. But beyond that, ⁓ something that I've actually been using in the past couple of weeks or so is this humanizer skill. So there's actually a GitHub skill where it's just going to humanize everything. So what it says specifically, it's just remove signs of AI generated writing from text used when editing or reviewing texts to make it sound more natural and human written. So if you guys pull up, yeah, if you pull up, ⁓ I think it's by Bader.

Corey Ganim: ⁓ that's awesome.

Nick Puru: B-A-D-E-R on GitHub. This is something that I have been using where it's just going to make it sound a little bit more, I guess, human for lack of better term.

Corey Ganim: And that's critical. I mean, that's something that even in, example, like in my open claw agent, one of the things that I have in its agents dot MD, which is like the open claw equivalent of claw.md is to never use ⁓ dashes ever, ever, ever. And it's still from time to time does. And I find myself having to go in and remove them because I don't want those in there. Cause people just assume that that's AI generated. So yeah, having a humanizer skill like that, I think is really helpful. just to do that kind of quick quality check and remove anything that looks blatantly AI generated. So that's a great tip there.

Nick Puru: Yeah, definitely. It's always the ⁓ dashes, the what's a few others. It's, yeah, bullet points or some overly formal structure. Exactly. Yeah. So that's exactly what this does. And I mean, that's not to say you have to use this specific GitHub skill. I mean, there's thousands of different alternatives to eliminate all that AI slop, I guess.

Corey Ganim: It's not X, it's Y. Right. Yeah. Yeah, there's, there's quite a few of, and there's plenty of like I see on X every single day, people sharing new skills or new repositories. mean, free resources for people to go and take advantage of to power up Claude code or their AI agents, or even just Claude chat and Claude cowork. So there's, there's so much out there that you can add on to the existing tools that you're already using. So this is fantastic. And again, just one more time to, you could kind of summarize. If somebody wants to get started with this inside of cloud code, building a, essentially a short form content creation factory, would you say the best place for them to start is kind of what you said earlier, just take inventory of what you have and then slowly build out the other context files and skills on top of what they already have.

Nick Puru: Yeah, most definitely. Some framing that I've kind of been pushing on to some of our recent engagements is, and even within some of my videos, really just trying to scream from the roof. Like this is essentially just another employee. So if you're going to be onboarding a new person, what does that actually look like? Well, you have to give them information. You have to give them context about the business, what your processes actually look like. That is exactly what you're doing with this AI system. So. ⁓ more specifically creating your Claude code environment. ⁓ creating the Claude MD. So finding out, taking inventory of any SOPs you guys currently have any context by your business, even if you just provide it with your website. And so literally going to Claude code or going to Claude, whatever, ⁓ tool you're actually using, you can just type out, okay, okay, here's my website. A little bit about us. And then, mean, one, it's going to do research your website. It's going to find plenty of information about what you guys do, who you are, who you've worked with, or you just provide it with that information. And then from there, really describe like what you're trying to get out of this. So you can create different cloud code projects. for example, this is short form. I have so many other ones. ⁓ So I have one that is for long form. I have a Hermosy consultant. I have one for creating LinkedIn posts. and anything for our B2B education side, I mean, the list goes on and on. So I'm just creating a different environment for each purpose. So if you want to create one that's for creating content just like this, just start with providing a context. Start with providing it like what you want to get out of it, things that you have learned in the past, or you could just have it researched extensively. So you can say something like, I need you to find the top short form video process for Instagram. So you have to research different communities. Don't bias to the first thing that you find and find generally what is the best approach that people take for creating short-form videos. So something like that. It's going to do a hell of a lot of research for you. It's going to compile everything that it needs. I actually took an approach. ⁓ doing something like this for LinkedIn and X and I've yet to actually put it into practice, but interested to see how that's going to end up turning out, but I guess I'll have to keep you in the loop within the next few weeks or so.

Corey Ganim: Well, and I think you nailed it, right? Talking about how important context is and looking at AI like it's an employee, because that's exactly what it is. And that's something I tell people is like, look, you wouldn't expect to hire a brand new employee and them to be perfect and knocking it out of the park on day one. Right. But on day 30 and day 60 and day 90, once you've trained them and given them your SOPs and taught, showed them about how you do business, they're going to do a great job. And it's the same way with AI. ⁓ and another thing I've said in my content in the past too, is that like the era of prompting it's over is over. We're in the era of context now and context is, what's most important. Like knowing how to prompt. They've, they've pretty much solved that. Like you no longer have to say like, Hey, act like a world-class copywriter or whatever the stuff that people used to say. It's like, just give it good context and a good clear prompt and it can go out and do exactly what you need it to do. Just like a really well-trained employee. So I think you're spot on with that.

Nick Puru: Yeah, most definitely. I think it just comes down to AI being put on a pedestal and all this hype on the internet. So everybody comes in with these expectations that like it's going to solve all your problems on day one when it's clearly not the case. yeah, I guess that's just what it comes with utilizing the internet.

Corey Ganim: Yeah, for sure. Well, Nick, this has been fantastic, man. So of course we're going to link your YouTube channel in the description and in the show notes, but aside from YouTube, do you want to send anybody anywhere to go and connect with you or work with you? Maybe if they need AI implementation in their business.

Nick Puru: Yeah, absolutely. So for any of you watching, any business owners, anyone ⁓ who's looking to upscale their team, educate their team, and really just become AI native within the next decade, then you can book in a free call. have ⁓ links all over all my platforms, YouTube, this is going to be posted, but you can check out inside of my bio. yeah, that being said, I'm pretty much across all different platforms, Nick Peru, so guys can check me out there.

Corey Ganim: Awesome. Well, Nick, thank you for your time, man. And be sure to go check out his channel guys. He's got some really good stuff on there and for everybody in the audience, we will be back soon. Thanks, Nick.

Nick Puru: Awesome, thanks Corey.