HSHrishith Savirindefgpt.hashnode.dev·Aug 22, 2025 · 3 min readAdvanced Rag Patterns And PipelinesRetrieval-Augmented Generation (RAG) has emerged as one of the most effective ways to bridge the gap between Large Language Models (LLMs). The basic RAG loop—retriever + generator—works however deploying it at scale and ensuring reliable, accurate ou...00
HSHrishith Savirindefgpt.hashnode.dev·Aug 20, 2025 · 1 min readWhere RAG FailsRetrieval-Augmented Generation (RAG) combines retrievers (that fetch relevant information) with generators (that produce natural language responses). But like any system, RAG is not foolproof. If not designed carefully, it can fail in subtle but frus...00
HSHrishith Savirindefgpt.hashnode.dev·Aug 20, 2025 · 2 min readINTRO TO RAGsGenerative AI models like GPT and LLaMA are great at producing human-like text, but they often suffer from LLM hallucination — confidently providing answers without actually verifying with real time actual data. That’s where Retrieval-Augmented Gener...00
HSHrishith Savirindefgpt.hashnode.dev·Aug 18, 2025 · 1 min readConcept of Agentic AI, explaining what agents are, how they work and the role of toolsAn AI agent is a system that doesn’t involve inherently in text to text response generations but interacts with it’s environment to achieve certain objectives - this is possible by giving access to internet to agents that allows them to surf through ...00
HSHrishith Savirindefgpt.hashnode.dev·Aug 15, 2025 · 1 min readBuilding Thinking Models with CoTChain-of-thought simply asks the models to show its reasoning before arriving to an answer that eliminates the “immediate jump to conclusion” step taken by most LLMs. This steps eliminates - “Hallucinations“ therefore breaking the problem statement i...00