Chetan Karnatakam — ArticlesGenerative AI in production: retrieval, agents, architecture and operations.https://www.chetankarnatakam.com/en-gbYour RAG isn't bad. Your chunking is.https://www.chetankarnatakam.com/articles/rag-is-not-your-problem/https://www.chetankarnatakam.com/articles/rag-is-not-your-problem/Most retrieval failures get blamed on the model. In production they almost always trace back to how the documents were split — and that is a much cheaper thing to fix.Thu, 10 Sep 2026 00:00:00 GMTRAGRetrievalProductionLLMAgent frameworks are a prototyping tool, not an architecturehttps://www.chetankarnatakam.com/articles/agent-frameworks-production/https://www.chetankarnatakam.com/articles/agent-frameworks-production/LangChain, LangGraph, AutoGen and the rest get you to a working demo fast. What they do not give you is an operable system — and the gap between those two things is where most agent projects stall.Fri, 28 Aug 2026 00:00:00 GMTAgentsLangGraphArchitectureProductionWhat teaching 15 engineers at a time taught me about AI systemshttps://www.chetankarnatakam.com/articles/what-training-taught-me/https://www.chetankarnatakam.com/articles/what-training-taught-me/Running live cohorts changed how I architect. When you watch people try to operate a system you designed, you find out very quickly which of your decisions were actually good.Fri, 14 Aug 2026 00:00:00 GMTTeachingMLOpsCareersHouse of Data