Vector databases
pgvector, Pinecone, Weaviate, Qdrant, Chroma, Turbopuffer, Milvus, LanceDB — picking a vector store for production RAG.
Document parsing
Turning PDFs, HTML, Word, slide decks, and scanned images into clean text and structure for RAG.
Synthetic data tools
Distilabel, Argilla, Lilac, Gretel, PromptWright, Bonito — generating training and eval data when you don't have enough logs.