Alright, everyone, welcome to uh the lab on vector databases. Today we're gonna be building a retrieval augmented generation system using my social media content, Pinecone and uh GPT 4. So, to start with, You want to make sure you have a pinecone account and log in, do the thing. And then when you're in here, you're gonna have like some indexes that you can create. So I'm gonna create an index here, and I'm gonna call this, or I was gonna call this the data expert, called the data expert index. So, this is an important part of like your configuration. Remember how I was talking about like Euclidean dot product, all those different ways that you can do your uh vector similarity. So, we're just gonna stick with cosine today. And then we're actually gonna use setup by model. And uh in this case, you'll see there's like 5 trillion different models that you can pick from. Uh, you'll see on the right here, there's like, if you want an image database or a text database, all those different things, it's all available here, but we're just gonna be going with the text embedding A02 model. So we hit set and then uh obviously there's like all this like, like shit around like whether you want to like server list or pods or whatever, like it all doesn't matter. We're just doing server list. I, I have not given these guys a dime, by the way, just nothing. So we're going to, uh, we're going to create. So, this is initializing our index. So, we have our index now.