Getting started with Pinecone (November 2025)

In this session, we'll briefly cover what Pinecone is, which use cases it serves best (hybrid & semantic search, recommenders, RAG), and why it's the leading vector database for building accurate, low-latency AI applications. Then we’ll demo hybrid search with Pinecone Database and build an LLM chatbot with high-quality RAG using Pinecone Assistant. We’ll wrap with Q&A.

Agenda

• Overview (5 min): What Pinecone is, core use cases (hybrid & semantic search, recommenders, RAG), and key advantages (accuracy, low latency, scale).
• Live demo (10 min): Hybrid search with Pinecone Database - create dense and sparse indexes (semantic + lexical), upsert vectors, query both dense and sparse, apply metadata filters, and use a reranker.
• Live demo (10 min): LLM w/ RAG using Pinecone Assistant - create assistant, set system prompt, upload knowledge base, prompt your assistant
• Q&A (5 min)