Set up the vector database pipeline using Pinecone. This involves creating a service that can:
- Take extracted text from a document.
- Split the text into smaller, meaningful chunks.
- Generate vector embeddings for each chunk using an LLM.
- Store these vectors in a Pinecone index for later retrieval.
Acceptance Criteria:
- The application successfully connects to the Pinecone service.
- A function can take a large block of text and store it as chunked vectors in Pinecone.
Set up the vector database pipeline using Pinecone. This involves creating a service that can:
Acceptance Criteria: