Your AI design team

0:00 And this automation allows us to leverage that at scale and also include images in our API request. So we cannot just make things from prompts, but we can give the API our own image that we wanted to edit or build off of. And so I think a great use case for something like this is like building images, building advertising materials in something like Amazon, right? So for my own example, my brother and I, we're in the process of creating our own like creatine gummy brand and one of the big players in the market right now is create.

0:46 And so when we've been talking about, hey, what do we want to do for our own advertising? the natural thing to do is go look at what the people who are number one in that category, go look what they're doing and do your best to like copy it or do something in that same vein. So, what you're able to do with this automation is take something like this for example, like this exact ad, go into Air Table, add it, and then add your own imagery.

1:12 In this case, this is our branding. and then create an automation that sends those images alongside a prompt that says like, "Hey, I want you to take my product and replace it in this image and then have it actually spit out a bunch of different variations of that." And so what you see here is exactly that, right? It replaced the create stuff with the max branding and gives me a bunch of different variations. And I can continue to iterate on that, right?

1:36 Like I could take these images, put it into Air Table as well, and give it more edits and more prompts. And I think this is a really cool automation because with Air Table, we allowed to have all this in a central location, right? I'm not just going into chat GBT, continuously dumping in images, having to scroll up and down the chat window to find what I did. No, this is all in one place. Super easy to figure out, hey, where do I put my prompt images?

1:59 Where do I put my prompt? What I want the size to be, the number of images, the quality, etc. And then all the endgame results are right here in the same place for you to compare. So, I think it's super cool and it's actually really easy to use. So, we're going to go through this step by step. I'm going to show you how it works from beginning to end and then we're going to go through each of the modules in the sub workflow so you can actually build this yourself on your own.

2:22 And so, what you're looking at is actually kind of a big picture overview of this workflow. Right up top here in the blue and the green, we have the orchestrator agent. So, this is where all the information from Air Table is going to be sent. And here inside the AI agent, it's going to decide to put it down one of two paths. You're either going to go down the edit image path or the create image path.

2:22 Now the edit image path is if we give the API images t to work with right like if we look here at air table when I have these prompt images right these these images I was working with that would be an edit image thing I'm saying hey take these two images look at them combine them do whatever that's the edit image path and then the other path which you see right here on the right is the create image path that's just a simple prompt to image generation workflow right I just give it a prompt saying I want you to create a b or c and it's going to go through this little workflow send the prompt open API create the image and it will populate back inside of air table as well.

3:08 So that's big picture how it works. Now let's actually dive into how the air table is set up and then we'll go through each of the modules one by one so you can see what's going on under the hood. But just as a quick plug for my school if you're