Already building a vector search stack (Qdrant + embeddings + RAG) and clearly invested in the problem, but no indication they're experiencing performance pain or actively shopping for managed solutions. if you want to check (Rivestack.io)