Not every AI feature needs an agent
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.

Series
Tested notes on building AI features in real products. Every article comes with code you can run: https://github.com/sayamjn/shipping-with-ai
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.

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

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
