SSSanjay Sorathiyainsanjaysorathiya.hashnode.dev·8h ago · 12 min readVectorless RAG: Can We Build RAG Without a Vector Database?Vector databases are one way to retrieve context for a language model. They are useful, widely adopted, and often a good default. They are not the definition of retrieval-augmented generation (RAG), t00
SSSanjay Sorathiyainsanjaysorathiya.hashnode.dev·6d ago · 3 min readThe Decision Layer Revolution: Why AI Apps Need a Decision LayerIn production AI software today, a surprisingly large percentage of LLM calls don't actually require generating open-ended text. Instead, they are making tiny, deterministic choices: Which team gets t00
SSSanjay Sorathiyainsanjaysorathiya.hashnode.dev·Jul 12 · 5 min readUnderstanding Retrieval-Augmented Generation (RAG) ArchitectureWhat RAG is and why it was introduced? imagine hiring an over-enthusiastic new employee who refuses to read any current news but still answers every question you ask with absolute, unwavering confiden00
SSSanjay Sorathiyainsanjaysorathiya.hashnode.dev·Jul 1 · 6 min readThe Complete Dummies’ Guide to LLMs chit chatConversational AI like ChatGPT feels magical: you type a question and get a relevant, fluent reply within seconds. Behind that instant reply lies a stack of ideas and algorithms that turn your words i00
