Getting Started with Pinecone

Datavolo helps data teams build multimodal data pipelines to support their organization’s AI initiatives. Every organization has their own private data that they need to incorporate into their AI apps, and a predominant pattern to do so has emerged, retrieval augmented generation, or RAG.

By leveraging RAG, you empower your AI application to furnish users with instant access to precise, up-to-date, and pertinent information. Whenever one of your users poses a query to your app, they receive responses honed on your business data, mitigating hallucinations and driving accuracy and relevance.

In our Datavolo Showcase, we have provided a blueprint for building a simple RAG flow, using Pinecone as the vector database. Pinecone is a great vector database to start with because of its compelling features for storing and searching data.

To get started with this blueprint, after you've downloaded the template and moved to your Runtime, you will need an account and API key to authenticate to Pinecone's API. When using Datavolo, you'll just need to sign up for an account and create a starter index. You can check out Pinecone's quick start guide here, but when using Datavolo, you'll only need to worry about account sign-up and getting your API Key:

With the starter index, you can use 512 for dimensions since we'll be using the text-embedding-3-small embedding model from OpenAI. You can also set the metric to cosine for cosine similarity.