🌲 Pinecone Flashcards
🌲 What is Pinecone?
Pinecone is a fully managed vector database for building fast, scalable, similarity search applications with ML/LLM embeddings.
⚙️ What makes Pinecone different?
Pinecone handles indexing, filtering, and retrieval out of the box with low latency and no ops, unlike traditional databases.
🧠 What data does Pinecone store?
Pinecone stores vector embeddings and metadata, enabling semantic search and hybrid filtering.
🚀 What are typical use cases?
Semantic search, product recommendations, RAG with LLMs, document retrieval, chatbots, anomaly detection.
📌 Is Pinecone open source?
No, Pinecone is a proprietary managed service. It provides SDKs and APIs but does not offer a self-hosted version.
🔄 What is hybrid search in Pinecone?
Hybrid search combines vector similarity with metadata filtering for more precise results.
🧱 What index types does Pinecone support?
Pinecone currently uses an approximate nearest neighbor (ANN) algorithm optimized for performance and scale.
🔐 Does Pinecone support security & privacy?
Yes. Pinecone offers encryption in transit & at rest, role-based access, and secure API key management.
💡 How does Pinecone integrate with LLMs?
LLMs use Pinecone to retrieve contextually relevant documents before generating responses (RAG pattern).
💸 Is there a free tier?
Yes, Pinecone offers a generous free tier with limited vector storage and queries, suitable for prototyping.