
RAG is a search problem first
When a RAG answer comes out wrong, the first fix most people reach for is the prompt, or a bigger model. Often neither can help, because the model never saw the text it needed. Hamel Husain and Shreya
Tested notes on building AI features in real products, by Sayam Jain, an AI Product Engineer in Bengaluru. Every article comes with code you can run. The originals live at https://www.sayamjain.com/blog

When a RAG answer comes out wrong, the first fix most people reach for is the prompt, or a bigger model. Often neither can help, because the model never saw the text it needed. Hamel Husain and Shreya
Here's how a lot of prompt changes ship. You tweak the prompt, try it on three examples you remember, the answers look better, and you ship it. That's vibes. I've done it too. Then I spent a few month

These days almost every LLM integration gets called an agent. Add a model, give it two tools, and suddenly it's "agentic". I build a WhatsApp assistant for myself, and it is deliberately not an agent.
