# Multi-Agents Are Out, PID Controllers Are In

DevFeed: [Multi-Agents Are Out, PID Controllers Are In](<https://devfeed.tech/articles/multi-agents-are-out-pid-controllers-are-in-33459.md>)

Original publisher: [Read original article](<https://timkellogg.me/blog/2025/03/06/pid-controllers>)

Published: 2025-03-06T15:00:00Z

Content type: opinion

Language: en

Sources: [Tim Kellogg](<https://devfeed.tech/sources/tim-kellogg.md>)

Topics: [multi-agents](<https://devfeed.tech/topics/multi-agents.md>), [systems](<https://devfeed.tech/topics/systems.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [llm](<https://devfeed.tech/tags/llm.md>), [multi-agents](<https://devfeed.tech/tags/multi-agents.md>), [systems](<https://devfeed.tech/tags/systems.md>)

## AI overview

An opinion article argues that many multi-agent systems become counterproductive because they add complexity through static roles and workflows. It presents a simplified Plan, Act, Verify loop as a PID-controller-like approach that can help keep long-running AI agent systems stable through small incremental changes.

## Source excerpt

My hottest take is that multi-agents are a broken concept and should be avoided at all cost. My only caveat is PID controllers; A multi agent system that does a 3-step process that looks something like **Plan, Act, Verify** in a loop. That can work. Everything else is a devious plan to sell dev tools.