i replaced an LLM classifier with twelve lines of if-statements and the client was happier
Summary
The author replaced an LLM classifier with a simple set of if-statements and found the client preferred the rule-based approach, highlighting the value of simplicity over complex AI.
Similar Articles
A client paid me to rip the AI out of the tool I built them.
A developer built an LLM-powered ticket routing tool, but the support team distrusted the black-box decisions. The client paid to replace the LLM with a simple rules engine, resulting in higher accuracy, lower costs, and greater user trust.
I built an AI trading agent, then ended up building one that doesn't use AI to trade
The author built an AI trading agent but found that using LLM for direct trading decisions is unstable, leading to a deterministic system where LLM translates strategies into explicit rules for execution.
My AI agent kept misreading my business logic. So I built a different way to pass it in.
The author built a browser-based editor for a methodology called Rulemapping to pass explicit business logic to AI agents, reducing misinterpretation by separating rule definition from execution.
Most 'AI agents' today are just if-else workflows with an LLM bolted on, not real agents
The article argues that most so-called AI agents are actually simple workflows with LLMs attached, lacking true adaptability, and provides a test to distinguish real agents from disguised workflows.
The most reliable data agent I've shipped is ~90% deterministic code. The LLM just parses intent and talks. Change my mind.
The author argues that the reliability of AI agents comes from deterministic code, not the LLM, and shares five key practices for building trustworthy agents on messy real-world data.