A hands-on video tutorial explaining how to take a RAG app from prototype to deployment using Pinecone serverless:
(1) Prototype a RAG chain with Pinecone serverless in notebook
(2) Use LangServe to convert the chain into a web service
(3) Deploy it using Hosted LangServe
We see demand for tools that bridge the gap between prototyping and production. With usage based pricing and support for “unlimited” scaling, Pinecone Serverless helps to address pain points with vectorstore productionization that we’ve seen from the community. This video builds a RAG chain that connects to a Pinecone Serverless index using LCEL, turns it into an a web service with LangServe, uses Hosted LangServe to deploy it, and uses LangSmith for app observability.