Leveraging AI

222 | Scrape, Analyze, Generate: Build Scalable Content Systems That Works with Nadia Privalikhina

• Isar Meitis, Nadia Privalikhina • Season 1 • Episode 222

Share episode

What if you could reverse engineer viral content and use AI to build your own content machine without writing a single line of code?

In this session, we go beyond theory and into execution. Step by step, you'll learn how to scrape top-performing YouTube content, analyze it using Gemini and ChatGPT, extract what works, and generate AI prompts that produce high-performing visuals and copy. All built with accessible tools like Make, Airtable, and ChatGPT API.

Our guest, Nadia Privalikhina, is not just a power user, she’s a systems thinker with a bias for action. With experience leading innovation and AI at scale, she's now building cutting-edge automation systems that blend marketing intuition with serious technical chops. Her content regularly turns heads and clicks on LinkedIn. This is your chance to see exactly how she does it.

Expect live demos. Real prompts. Actual outputs. And the exact workflow behind a system that can transform how your business creates content.

About Leveraging AI

If you’ve enjoyed or benefited from some of the insights of this episode, leave us a five-star review on your favorite podcast platform, and let us know what you learned, found helpful, or liked most about this show!

Isar Meitis:

Hello and welcome to another live episode of the Leveraging AI Podcast, a podcast that shares practical, ethical ways to leverage AI to improve efficiency. Grow your business and advance your career. This is Isar Metis, your host, and if you are listening to this podcast, it means you are consuming content that somebody else generates in this particular case, me. Uh, but generating content in general is a great way to build an audience and create relationships. And if you nurture these relationships and you continuously provide them value, it will over time lead to business. I've been a big believer in that for many years when I work for large companies and now when I am doing, my thing right now on my own. But either way, creating valuable content that really helps a specific audience, that is your target audience, is a great way to grow your business. However, knowing. Exactly what to post and how you should post it is not that easy. Many organizations and many people have been running very, very fast on the content treadmill, but going nowhere because they're not necessarily sharing content that helps their target audience in a meaningful way. But to do that properly, it's actually a lot of work, right? You gotta do research and figure out what's actually connecting with these people. You gotta do research and figure out what's trending right now and what they might be interested in. You gotta then analyze that content and figure out which components they might be interested in, and only then you need to start generating the content and generating the images and all. It's a lot of work, and so. Doing that manually is something that many companies and definitely individuals are not doing, or at least not doing effectively. However, AI is a great platform that knows how to do all these things. It knows how to do research, it knows how to analyze data, it knows how to create content. So if you can take each and every one of these components with the right prompts and then combine them together into a single automated process, you can create an incredibly effective content machine, which is exactly what we're going to show you how to do today. Now, our guest today. Nadia Priva is an expert in AI automation. She's been building AI automations for multiple clients in the past few years. Now, her background is being a software engineer, which means she definitely understands the technical side and how to build things that flow effectively with minimal interference and issues. But she also spent a while running an e-commerce business, which means she definitely understands what people, what motivates people to buy and how to connect with them in order to drive them to take actions, which this combination makes her the perfect person to show us exactly how to do this process. Now, when I saw her post about this on LinkedIn, I'm like, oh my God, this is absolutely brilliant. I need to bring her on the show. So here we are, and I'm personally very excited, uh, for this episode. I know it's pure gold. Again, you will see how to do all the things we've just talked about in an automated way. Leveraging AI together with automation capabilities. And so, Nadia, welcome to leveraging ai.

Nadia:

Thank you so much for such a nice introduction. Uh, today we are going to talk about how to create YouTube thumbnails specifically that is the use case that we are covering, but I believe the same concept can be applied to different, uh, industries and yeah, so the

Isar Meitis:

automation that Nadia showed is how to take what's working for other people on LinkedIn. Combine it with the stuff that you want to include in your thumbnails, and then create thumbnails on the fly for your content. The same thing can be broadened to anything that you want to create, by researching, analyzing, and then creating based on the same exact parameters. you'll see it, it feels like magic once you'll see the results. I, for all of you who are joining us live, whether you are joining us live on LinkedIn or whether you are joining us live on, the Zoom call, then feel free to ask any questions. First of all, introduce yourself right now in the chat. Say where you're from, uh, say where, who you work for. Uh, send your, your LinkedIn link if you are on the Zoom chat so other people can connect with you and tell us in two words what is. Your biggest interest in this, uh, in this particular, uh, use case. And, if you're not here with us live, the question is why aren't you here with us live? We do this every single Thursday, uh, noon Eastern time. We have amazing people like Nadia that are gonna share exactly how to do different use cases. And then you can, uh, ask questions, which is the last thing that I'm going to say, and then I'm gonna give it back to Nadia. feel free to ask questions in the chat on LinkedIn or in the Zoom chat, and I will bring it up to Nadia, and we're gonna answer your questions with that. Nadia, uh, the stage is yours. walk us through your magic.

Nadia:

Yeah, thank you. So I think that I will share my screen because it is the best way to showcase what is what we are going to talk about. for those

Isar Meitis:

of you who don't see the screen, by the way, because you're listening to the podcast after the fact, then uh, we're gonna explain everything that's on the screen, uh, so you can follow us as long, Despite the fact that you're driving or walking your dog or on the treadmill or whatever it is that you're doing. However, if you have the opportunity to also watch the YouTube, then that's another great benefit because you can see afterwards what we've done.

Nadia:

Okay. And right now I'm sharing the screen with my LinkedIn post that's attracted that attention. And what happened is that I, I am part of Liam Soley, private group, uh, where we learn how to use AI and how to sell AI services to businesses. But actually, Liam runs a big channel with more than half a million subscribers. And one of the problems that, uh, he faced is that they need to constantly create thumbnails. They do research, they create those, uh, different variations and then post those variations. Yeah. So, uh, there was a hackathon and I created the system. I will quickly. Uh, run through it right now, but here you can see the first examples of what types of thumbnails the system produced. And yeah, I haven't mentioned that. I won that hackathon. I, took the first place, uh, with those images, and since then, a lot of people reached out saying that those are amazing, because if you don't know how Liam, uh, looks, he looks exactly like, uh, those images that I'm sharing right now to those who are listening, you have to, and that

Isar Meitis:

was pre, pre nano banana, right? So now, now this is probably even easier.

Nadia:

I would still say that, uh, character consistency can be a problem. So I tried after Nana Banana, uh, released to do the same visit, but the character consistency is still a problem. So I will tell you how I approached, this problem awesome. And how I solved it.

Isar Meitis:

So again, those of you're not seeing, there are multiple thumbnails with a very clear style. So they have these kind of like beige background or, or kinda like off-white background with graphics that are all look the same as far as like sketches and images of the person in multiple directions and angles. But his face is in most of these, most of these thumbnails and they look very consistent from a style perspective. And they look highly professional and yet no person has created them, which is the whole magic here.

Nadia:

And maybe even share a few of examples a little bit closer. So here are those finished thumbnails and how amazing. Okay. Lemme exit this screen, and now I will tell you how I approach this problem and how you can do this. Awesome. So the first, just a

Isar Meitis:

quick question. How long was the hackathon? Like, how much time did you have to work on this?

Nadia:

Well, in theory we had al almost one month, but, uh, as, because I have my clients and, uh, I believe e everyone else faced the same problem. So we finished it, in around three or four days. Oh wow. Okay, cool.

Isar Meitis:

Yeah.

Nadia:

Okay. And now I will share. So to prepare this and how would you create a thumbnail or any other, content, you would first want to understand what works and what works for others. And this is what I also did. So now you can see on the screen this. So I give it a moment and it'll display the thumbnails. I screen all the thumbnails from multiple creators on YouTube and created, um, my database of we need thumbnails. And later, initially, my idea was to analyze those thumbnails because you have models that can, transform visuals into text, for example. And then my idea was to train to fine tune a model, but actually it didn't work even though I used that fine tuning for my own LinkedIn content. Interesting. But yeah, but in this case, with someone else, for some reason it didn't work. So what I ended up doing is I created, I scraped multiple channels. Obviously there is still some human work. I had to pick those channels on my own. and then the system could, let me show you the, what. Going behind the scenes of that scraping part. So now I'm sharing my NI 10 workflow. It is super easy and quick, and someone mentioned that it's probably not legal to scrape YouTube, but actually YouTube gives you quite a lot of data points, including thumbnails that you can get absolutely, legally, uh, with some limitations. Interesting. So first I set the channels that I want to monitor, and then for each of those channels, I get all of the videos from the past year of them. And you can see when you're saying get the video,

Isar Meitis:

you're getting the thumbnail and the description.

Nadia:

let me show what we have here. We can, uh, YouTube provides a lot of information about those videos. So you can see we have title, we have channel name. It is somewhere description. Then thumbnails. Yeah, thumbnails here. And this and this

Isar Meitis:

node in NA 10 is just, that's what it does. It just grabs information from specific channels.

Nadia:

Yes. So it takes the channel id and I say that I want to get all the data from past year in this example, and then I get all of that video. So it found, I believe I entered around 10 channels and almost every channel posted more than 100 videos. So I got all of those 100 videos for each channel, which, makes it up to 1000 items in this case. Okay. And later obviously want to filter them. And yeah, there is another, data point I will not be sharing today because it'll expose my api. But there is another, option to get, um, more sta stats about that videos. um, if you count and comments and, um. Whatever you want. I would say there is still a lot of information that can be, and

Isar Meitis:

that's, and that's from what tool did you use that? Is that like an appify? Because it, see

Nadia:

no, it is also YouTube's official. Oh, so it's,

Isar Meitis:

it's YouTube's own on, uh, resources. Okay, cool.

Nadia:

Yes. Yeah, I just, and doesn't, uh, give you an option to list all of those. So in this case, I use

Isar Meitis:

Yeah. So to, to explain to people how these work. So all these tools work the same, whether you're using NA 10 or Make or Zapier or, or any other, they, in all their tools, they don't expose every single function of the API that exists in the tools they connect to, in this particular case, YouTube, which means you can. You can connect to the simple node and it will give you several functions that you can do in the node. But you can use an HTTP call, which is very, very easy. And you don't have to know how to code. You just have to get the documentation and drop it into a, your favorite large language model and say, I want to connect to that with an HTT P call and I wanna find that information. And it will give you the code and you can drop it here on N eight 10. and then you can call the rest of the functions that did not exist and was not prebuilt into the tool itself.

Nadia:

Awesome explanation. Thank you so much for that. And after we get that information, in my case, I wanted to filter out short form videos, but for some people it may be the opposite case, so they would want to see what's stringent for, uh, short form videos. And I had to use a little bit of code for that, but I believe that it's also possible to do without code, but because I have technical background, for me it was easier to use code in this. Then I just filter out, uh, short videos that are, uh, shorter than two minutes. And after that I get also, because we scraped all the videos from the past year, we can get a data about what is the average view for this specific channel and, uh, what is their average a like rate. And this, uh, will help us to identify what is, what are the outliers for this channel. Oh, cool. So in my,

Isar Meitis:

I see where this is going.

Nadia:

Yeah. It's, it's not that simple. I mean, it, it is, uh, quite simple, but you need to have your own strategy. This is where the human part comes in, and. So yeah, we get the average, views per hour for this channel and average lags per video. And then this information helps us to identify which videos are good and which videos are not that good. And after that, I use that information and save it to my data. I will just, and again,

Isar Meitis:

to explain to people, when you say database, people are like, oh my God, I don't know any databases. It just saves information to Airtable and those of you haven't used Airtable. Airtable is like, uh, Excel with, uh, better and easier to use user interface, right? So it allows you to do more filtering and control more, uh, the look and feel and add images. Like you see in the example here, you see all the thumbnails from all the different channels, uh, that this process has brought, but it, it's saved into Airtable, which is a very user-friendly tool. You don't need to know how to run databases or set them up or so on. You literally go to Airtable, create a new space in Airtable, define what you want to have in it, is it images? And what's gonna be the columns and what fields do you want in it? And then you just can feed that from whatever source. In this particular case, from N 10.

Nadia:

Absolutely. I have non-technical clients who are quite, uh, quite navigating, irritable very easily.

Isar Meitis:

Yeah.

Nadia:

And so I save those, uh, thumbnails. And then I have one more, uh, workflow that actually analyzes those thumbnails. So I use an agent in this case, but why I use agent, maybe it is not so specific and you can use, let's say your open AI can analyze what is on that image. But in my case, I decided that I will do this with agents. It's just a small limitation of an item that I found or what I wanted to use, that I had to pass the image and use an agent for that. So my agent analyzes those thumbnails and saves, uh, them to, again, to an air table.

Isar Meitis:

Can you explain what exactly it analyzes? Like what is it looking for?

Nadia:

Okay, let me see. So the prompt says that you are a world class YouTube strategist with attention to detail. You, uh, leave, and breathe YouTube content marketing strategy packaging, and are scaled at creating v packaging for high performing YouTube videos. You always output whether Jason, this is one trick that this, we just needed more structured output from this tool, but it's not mandatory, I would say in this specific case. And then I give an additional standard prompt where I say that, um, this agent will be given a YouTube video that performed well. And, uh, okay. I also feed it with a transcript, uh, title and thumbnail. And the mission of this agent is to deconstruct why this, uh, thumbnail works for the video. And I identify the thumbnail archetype and this.

Isar Meitis:

before you close this again to people. Uh, so what is JSON and how does it work? So, JSON is a very simple coding language that basically describes the data as if it is a table. So it tells you which fields they are and what values they can have. And the benefit of using JSON is that more or less every tool out there knows how to use it. And so if you export and or import that, then you get consistent results because it's, it's gonna come in a. Very well-defined format. And again, you don't need to know what it is. You can go to whichever tool and said, I need a js ON that will gimme these parameters and it will write the code for you and it can paste it in here. The benefit of that, once the data gets into an automation tool like NA 10 or make or whatever, you can identify each and every one of the fields. You can map them to the columns in the table in Airtable and so on. Uh, so it just gives you a structured output versus just free text.

Nadia:

Exactly. And, uh, maybe I will. Okay, let's move on. And so this system could, uh, look at all the thumbnails and transform them into, text representation. Okay. And then I say that information again to Airtable. And so, uh, by the end of that second workflow, what I have is a lot of structured data. I have titles of, Highly performing videos, channel names, summary and then thumbnail description. And for thumbnail description it says what is in the background, what is the text on the thumbnail? And, um, a few other things that we will, yeah, what is, on the foreground? And those are the main components. And so you basically

Isar Meitis:

deconstructing the thumbnail into its various components. And because these are all the top performing, uh, the top performing videos from the specific channels, then that hints that these are solid descriptions and or thumbnails or a combination of the two, which gives you hints to what you need to do in order to create successful ones yourself.

Nadia:

Yes. Now we are just preparing the data for, uh, the future automation. Yeah. Awesome. This is the data preparation step. What we do now, Airtable allows you to export that data. I don't remember where it is, but you can actually export all the table into a CS file and what we do next. And you can do it with any types of content. I did the same with LinkedIn content. So you export it, and then in my case, I went to Google AI studio, or you can go to Gemini, but Google AI Studio is a similar version. So it is just Gemini for developers where you can safely test your ideas and experiment with, that model and how it works. So what I did is I went this Google AI studio and I exported, and I give it that CS file with description of best performing videos alongside with, uh, the thumbnails because we have. already transformed the, uh, the visual thumbnails into text. And so Jam and I can work now with that text. And why Jam and I, because it has a huge context window. It has more than 1 million token, a context window, which means that you can basically feed it with, in my case there was around 100 videos, but you can potentially give it 2000 videos, for analysis. And so I, after that, after I give, uh, it this CC file, I instructed that, uh, you now see perform, top performing thumbnails for videos, their description, explanation, and so on. And your mission now is to generate a prompt for an LLM. This is also a secret trick that I use. I always ask Gemini to create prompts for me because it is just, it knows how to create prompts better than me. And also I give it a link to the, uh, prompting guide for gem. Google publishes those, prompting guides. They are officially available. So I give this prompting guide, uh, along with the Cs, a file. And now what Jim and I did on this step is that it came up with a good prompt. It now created a structural prompt with a persona. So again, it, uh, says that you are world class YouTube strategist, and it is not my prompt already here. This is what Jim and I created. Yeah, and it created the task. So task is to analyze and provide video title and summary, and generate, uh, in this case, in the context I mentioned that I want to generate, based on the title of the video and transcription of the video, I want to generate those thumbnails. And so all that information, came into the prompt. Later I made a few modification to inject the branding, guidelines that I wanted it to have, but the structure is the same. Uh, I will just iterate upon the first version and ask to add a few more components to it. Okay. I believe this is the second one. Yeah. I ask it to deconstruct and reverse engineer how I can get a clear description of a thumbnail based on title and transcript. And I used one of those prompts, uh, later on in inside and automation.

Isar Meitis:

So the prompt, again, just to understand what it does, it takes all the inputs from the CSV file that was created based on all the data that we collected previously, and now it gets an input for a new video.

Nadia:

I ask it now to create, prompt for basically itself. I say that it'll be given yes. Title and transcription of the video. And

Isar Meitis:

it needs to then define what the exact, yeah. Okay. What the exact sum needs to be based on the best practices that it learned from all the previous material. Combining it with the new need for the new, uh, video,

Nadia:

something like that. Yes.

Isar Meitis:

Okay.

Nadia:

And then, uh, is there a

Isar Meitis:

section for like brand guidelines and something like that? Because again, we saw everything was perfectly branded when you showed us.

Nadia:

Yes. Later I injected it, manually or Got it.

Isar Meitis:

So, so you added manually a segment about the, the branding of it. Exactly,

Nadia:

yes.

Isar Meitis:

Yeah.

Nadia:

And after this step, I have a prompt that I can use, but let's go back to the, this is. Now on the screen, you see the final automation that I created. It starts with a form. So we need to somehow instruct the system that, hey, this is our title, and hey, this is our, what is our video, about. And so it starts with a form where we feel that information and later it can summarize the transcript if it's needed. And, then I use that exact prompt from Gemini. It has this persona about world class, YouTube thumbnail designer and task and instructions and so on. And then it comes up with, a few concepts. We don't want to have only one, uh, thumbnail because most of the channels test multiple thumbnails, and we want this system to create multiple concepts. So it came, uh, it comes up with a few concepts, and it is based on, uh, concept archetypes and so on. And. It outputs visual descriptions of what is, what should be in the background, what should be the primary subject? Is it a person or is it an just an object, which fits this video and what are the elements if there are some arrows or additional, I dunno, a pile of money or something to, evoke some emotions And then text elements. This is all what we need to almost all what we need to generate and visualize those concepts. Uh, but what else do we need? If it is a personal brand, then we obviously want to have a person on those thumbnails. And if you don't know, then um, open AI has this image model which allows you to reference images. So now we have none of banana Before we didn't have it, uh, but still open ai. Has that, possibility to reference not only one image, but multiple images. And it also can be done via API, which means that it can be automated. Yep. Now, on the screen, you see one of the examples that OpenAI, uh, provides, uh, there are four objects, and then if you give it a prompt to, compile those objects into one image, then it combines all of them and puts them into one image. And we use a similar approach with thumbnails. Okay. So what I did here is I used a few, thumbnail examples from the channel. Okay, let me open it. So here is one example and there are a few more. So now I use those three examples and then, There are just a few support and blocks, but later I ask an open air model. So I reference those images, multiple three images, and then I say that again to that. It is, an expert YouTube art director and a master of photo realistic concepts executions using generat, FAI. Its sole mission is to create a single masterpiece level YouTube thumbnail based on the detailed concept provided below. Uh, you must perfectly, use the visual style guide and the perfect and specific concept into one adhesive hyperrealistic image and so on. And here I also give it additional, references. So I mentioned that the images that I provided are the reference of a creator of this channel, and I also give additional styling guidelines. Right here in the prompt. So we keep styling guidelines into, multiple places because the first place it affects the layout of the image. And in the second place when we actually generate that image, then yeah. styling guidelines affect how it looks and what colors Yeah, it use. And so it, let's actually use, so this is one, image, another input image, and then we give it a concept. And finally, I will not show it, right here, but let's go here. okay. This is,

Isar Meitis:

so these are few. So again, what, what you're not seeing is that basically what this step does in the automation is it takes. Basically everything we've prepared so far, but mostly it cre It has a prompt on exactly what to include in several different variations of the thumbnail. So it, it take, it takes these, uh, as NAIA called of archetypes of different, what works, basically. So it's gonna generate several different versions to use. It's using images from previous thumbnails to get a visual reference of the brand guidelines as well as the look of the person. And, uh, then it combines all of that into outputs. So every time you run this, you're gonna get x, I dunno how many, three, four, uh, different options, four thumbnails that are already aligned with the brand, that already has the person, the text, the background, the uh, supporting graphics, whatever needs to be in there already in the image.

Nadia:

Yeah, exactly. So we first distilled what is working into a prompt. Then that prompt created a few variations, a few text representations of, variations. Yeah. And then we kind of converted that text into an actual image together with a few reference images that we use them only to reference, base of a creator. That should be on the thumbnail. Uh, but if you are watching it, you see that the final images, they are not perfect. So even though we give charge GPT, the model behind charge GPT, reference images, it doesn't, uh, preserve that character so well. And that is why we need additional steps. And anyway, this image is not in the proper format and so on. So we want to do the face swap. And face swap was one of the, um, hardest things to figure out. But what we found is that there is this flux context model. You probably know about it before, again, before nano banana flux context was one of the best models for visual editing. And with it you can modify some text on a, an existing image, you can trans, uh, transform that image into a different style. And one good thing about it is that you can actually create a fine tuned version of this black context. And what it allows to do, let's say we have an image with not a perfect face. Let's say we have, uh, this image and we want to use, we want to have an image with an accurate face. So we would train a model that has an input image with, uh, someone else's face. And the final image would be the face that we want to see. And in this way, we can later on. Run that fine tuned model on our, on the images that our LGBT model produced. I hope it makes sense. Interesting.

Isar Meitis:

Yeah, so I'll, I'll, I'll re let me explain this in a minute. So, first of all, flux is an open source model, uh, that generates images and can edit images. Probably the best open source model that there is out there right now. Uh, they've been around for a while and they became extremely, uh, like they created a really big buzz because it was really easy to train them. They're called Flex Loras, and you can literally just take, uh, it's a relatively simple process. Many tools allow you to do it outta the box. Uh, but you can just, upload multiple images that are examples, of either how you want the image to be, or like Nadia said, examples of before and after. This is my before. This is my after. This is my before. This is my after. This is my, before this, my after. You load as many of these as possible and then, or not as many as possible. Like you, you need just a reasonable amount. Like one or two is not enough, but 15 to 20 is definitely enough. And then what you can do is if you have a before and after all the afters showing the correct face, then it learns how to do that. And then you give it a before image and it knows how to change it to an after image. I actually haven't seen anybody use it for this kind of use case. So I found that you're doing absolutely brilliant. I think this is so freaking cool.

Nadia:

Thank you. because it appeared not so long ago, around a month or two ago, I didn't find many tutorials that I talk about it. So it was quite challenging to find, to figure, to figure it out on your own. Yes. And here you can see one example. So this was the, input image and the face here is what GBT produced. It is quite good, but still the face is not the face of the person that we want to see. And after running this model, it produced a much more, familiar face to

Isar Meitis:

do you have. So just as an, uh, question, uh, for the audience, do, is there a, do you also train it just on the face separately or All the images are just before and after with him? with a good image and a bad image.

Nadia:

So you need to have, um, obviously we need to have a photo shoot or existing thumbnails. You can take them from that database, that from the initial steps. When you scrape those thumbnails, you can use them as a final image. The problem here is to create the initial image. Something that,

Isar Meitis:

yeah, to, to create the before images. So how did you, that was my next question. Again, to explain my, my question in the dataset, when you're training it, you need a good image, which, okay, you have the guy's YouTube channel, you have his existing, his existing thumbnails. These are easy, but how do you get the one that is not his face? you have to kind of like destroy or ruin the good images in order to do that. How do you do that?

Nadia:

The old way would be to use a Photoshop. The new way is to use the exact same model, flux context. Laura, or not Laura, just flux context. Or now you can use nano banana. You would give, uh, this model the final image and ask it to come up with another, just swap the face with someone else. Oh, see the face? See what you're saying? I see. Yeah. And this way your final image becomes your starting image for the training. Yeah, yeah, yeah,

Isar Meitis:

yeah.

Nadia:

And this is how we've done, so we have after this step, the. Thumbnails is almost, perfect. The Romanian steps are to just resize it or to feed YouTube's, nine, nine by six 16 or 16 by nine, uh, ratio. And also to, uh, upscale that image. But that is quite, so two questions about

Isar Meitis:

this. The resizing is done with what tool?

Nadia:

Resizing. Let me see. I'm okay because we, what charge GPT model open AI image model produces is, three by two, uh, size image. And we don't want to lose, some of the. Top how to explain top and bottom pixels on the image. That is why what I use is out painting. And you can use other models for out painting. What, uh, the platform that I used for, flux training is called file.ai. Yeah. And it hosts a lot of visual models as well. So you would go there and just find a model that I perform that can do out painting. I used, what did I use? I used some ID for out painting. Oh,

Isar Meitis:

interesting. Yeah.

Nadia:

And then I also found a model for upscaling.

Isar Meitis:

Which one did you use for Upscale? That's another interesting one.

Nadia:

that was a basic model from cloud ai. I don't even remember its name. It's, uh, it's not something that people talk about.

Isar Meitis:

yeah, yeah. There, there are, I, I actually have. One or two very good open source ones that, that do a very good job in upscaling. So I was, I'm always curious to see what people are using.

Nadia:

Yeah, you can, you can do it definitely on your local machine.

Isar Meitis:

No, I'm, I'm running it on file and I'm paying them either that or ate or one of those, like it's running somewhere. It doesn't run on my computer. Mm-hmm. But, uh, I see, very, very cool. So this is, yeah, I wanna run through a quick recap so people understand of the entire process. and then I will see if, if there's any question. Uh, so first of all, there's a question. how long did it take for this process to take developed? So we, we talked about this GW in the beginning. Maybe you missed that. it, it took a few days. uh, to create this thing and I'm sure, while doing other stuff, so I, I assume you didn't just sit and do this for a few days. I'm sure you have other stuff to do other than developing the automation. Is that a, is that a true statement?

Nadia:

Yeah, that's true. I spent a few. Nights developing it. Yeah.

Isar Meitis:

Yeah. Okay. So, so probably overall from working hours, were we talking, I don't know, 10 to 15 hours-ish?

Nadia:

maybe a little bit more than that. Maybe a little bit more that My nights long. Okay,

Isar Meitis:

cool. Long nights. A few. Long nights. Long nights. Awesome. Yeah. Yeah, yeah. It's, it's brilliant. So let's do a quick recap of what this entire process does and why is it so amazing. And then I'm gonna mention a little bit how you can generalize it. The way it starts is it starts with research, right? So Nadia started manually saying, okay, who has successful YouTube channels?'cause that's already gonna be a good place to start. And she randomly picked who have successful YouTube channels, and then she used inside of NA 10 A scraper to bring all the thumbnails of their information and then choose another call to YouTube to get additional information, such as how many views they have, which is, is a critical aspect to the next step. And then the next step was, filtering only the ones that are positive outliers, right? So if the average person has 10,000 views to each, to each episode, some episodes suddenly have a hundred thousand, which mean it caught people's attention. and, and by the way, the numbers, the actual absolute numbers don't matter. Like if the regular. View is 500, and then suddenly you have something with 3000. That's a good outlier. So then not only you're starting with people who know what they're doing because their channels are doing well, you're picking the ones that were really doing well from that channel, which means either the description or the thumbnail or the, like the topic, something in there, was working very, very well. And we use that as a way to train the AI on what is working from a thumbnail description, title, perspective on LinkedIn. So that just sets the stage that builds the machine. Then the other half is saying, okay, I want to create a new YouTube video about topic X. What should I put in the description? What should be the thumbnail? What should be written on the thumbnail, and so on. And so Nadia asked Gemini to create a prompt. On how to do that process. She used that prompt in the process. And what that prompt generates is it knows how to take the following inputs. It takes the outputs of the previous step, what is working and what's not working, best practices, and it's getting the input from the user on what the topic of the new podcast is, and it generates a prompt on how to create basically a very detailed description of the thumbnail. Then that gets fed into an image generation tool that actually generates the actual thumbnail. Then the next step actually swaps the person's face. To a better version of that face of the person, uh, using a trained AI model from FAL. So using flux context flux, again, is an open source model that you can train very easily just by uploading examples to it, and then that makes the face better. And then the final step was to get the right aspect ratio and the right resolution because in many cases, uh, the image generation tools generate really relatively low resolution images and in standard, outputs as far as aspect ratios. Some of them actually know how to generate the image in the correct aspect ratio to begin with, and some of them don't. So if you found one that gives you the right quality of thumbnails, but necessarily the right aspect ratio, then you can then change that, uh, as well. And again, just to. Dive a little deeper instead of cropping the image, which means you lose something and you never know what you're losing because you're not creating the thumbnails on your own. Uh, that did it the other way around. She used out painting, which basically generates new pixels around the image in order to extend it to whatever the aspect ratio needs to be. Um, again, absolutely mind blowing and brilliant. I'm not surprised you won, first place. So there's a question. It's a general question, but I think it's a good question to add to this. Uh, they're asking, are you using the cloud NA 10 or yourself hosts NA 10?

Nadia:

I self host on a 10. In the cloud if it makes sense. Yeah, yeah, yeah. I'm not hosting it locally.

Isar Meitis:

Yeah, yeah, yeah. So, so to, to explain the question to those of you who're listening, you can go to n eight ten.com and just sign up and then use it like any other software, right? And then you're gonna pay for usage. So the more automations you run, the more money it's going to cost you. Option number two, LA 10 is an open source model. You can go and install it. Wherever you want, like on a third party cloud, which is the way I run it. I run it on a hosting platform called Railway. Uh, I think most people who use NA 10 a lot host it somewhere and not use NA ten.com. Uh, it gives you several different benefits. Benefit number one is money like you for the same cost of costing of hosting. I'm paying$6 a month, I can run as many automations as I want. It doesn't matter. Uh, so that's a big benefit. The other big benefit is from a data security perspective. the data doesn't go to a third party server where you don't know where anything runs their data. It just stays on your server. And if you run it on a local machine, then. A hundred percent the data doesn't go anywhere because it just stays on your machine. So these are the two big benefits of running it this way. It sounds fancy, like how, I don't even know what open source is and I dunno how to host something on a third party server. Again, I just was looking for the easiest way to do this. The easiest way I found is railway, there's a dropdown menu. Say create an NA 10 instance for me. You click go and that's it, and you have one. So the knowledge you need is exactly zero, and you can have as many instances of NA 10, as you want.

Nadia:

I'm also using railway. I just wanted to add that when you work with images and videos, then like your server, server will cons, consume a lot of, memory. So you need to also keep it in mind. And because I, automate those images and videos quite frequently, then my usage cost goes up.

Isar Meitis:

Yeah. uh, but again, it doesn't go up. With usage, it goes up once to the level you need and then you can run it a thousand times. It's not gonna cost you significantly more money. Right.

Nadia:

It's, uh, you need to keep an eye on because stores your executions. Yes. And when you create multiple images or videos, then those, oh, executions are hosted for 30 days. They stay. Uh, so do you have

Isar Meitis:

a step that then deletes them at the end?

Nadia:

no.

Isar Meitis:

You do not. Yeah, because that's an interesting thing I'm doing in other automations. Now, now we're drifting a little bit, but in some of my automations, like in, in, in cases where I upload information to, uh, let's say vector stores of a custom GPT, then the final step of the automation goes and deletes the file. After I already have the output that I wanted, I go and delete the memory because otherwise again, you got, you gotta start accumulating, more and more stuff. But, okay. So. Nadia, this was absolutely amazing. Again, this is such a brilliant use case. And now I said I will generalize it for a minute. Think about what we've learned today from Nadia. We've learned how to research and scrape data to learn something about it. We learn how to filter it to find all in the stuff that is relevant and that is the best performing. We learned how to turn that into a prompt that can generate new variations of the winning concepts. And then we've learned how to apply that in order to create both texts and graphics that will mimic the successful thing. You can take that to anything from ads to posts, to, post on social media to blog posts. Like literally any content that you wanna generate can follow the same exact process that Nadia was showing. And then all you need is some. Basic, well, not basic. You need solid NA 10 skills to put it all together. but once you put it together, once you have a machine that can do it day in, day out, nonstop, which makes it really brilliant. Uh, this was amazing. Thank you so much. If people want to follow you, connect with you, work with you, learn from you, what are the best ways to do that?

Nadia:

Thank you. Sorry so much for, outside and how this all works because you explained it much better than me. I feel I'm on the technical side anyway, if Yeah. For people to find me. I, I am active on LinkedIn. It's just my name, NA and I also active on YouTube. I post long and I attend tutorials about video and generation as well. It is, na ai insiders.

Isar Meitis:

Awesome. Uh, and I wanna thank everybody who joined us live. both on LinkedIn and on Zoom. I know you have other stuff that you can do on Thursdays at, uh, noon Eastern. Uh, and I appreciate you being here. I appreciate you asking questions and introducing yourself and chatting, uh, in the chat. Uh, so thanks everyone. Uh, for those of you who are listening to this after the fact, come join us next Thursday. It's, uh, noon eastern every Thursday. You can join us either on Zoom or on LinkedIn, and we share the magic, right? It's brilliant people like Nadia who's gonna tell you exactly how to do really, really cool stuff, uh, with AI that you can start implementing immediately afterwards. That's it. Everybody have an awesome rest of your day.

People on this episode

Head Shot

Isar Meitis

Host