AI / LLMs
6 min read
Vectorless RAG: when you do not need embeddings
Vector databases became the default answer for retrieval. For a lot of enterprise knowledge work, they are more machinery than the problem deserves.
Read