Retrieval pipelines, chunking, embeddings, reranking and document parsing.
Document-to-RAG pipeline in a box, from a team that runs it at Tencent scale.
The unglamorous parsing layer that decides whether your RAG works at all.
Data framework for ingestion, indexing and retrieval — strongest on document-heavy RAG.
Turns PDFs, HTML and scans into clean structured elements your retriever can actually use.
End-to-end RAG engine with template-based chunking you can show to a customer.
Fine-tuning encodes behaviour; retrieval supplies facts. Asking which is better is asking the wrong question — ask how often the underlying knowledge changes and whether you must…