What are multi-agent AI workflows, and are they better than one agent?
Short answer
A multi-agent workflow has a coordinator agent hand pieces of a job to sub-agents running at the same time, like one researching, one writing and one assembling. In my tests it was faster and less error-prone than one agent doing everything. I got a local newsletter to roughly 97% quality in about an hour this way.
The day I tested Claude's dynamic workflows, my local newsletter production changed. Claude automatically created and coordinated sub-agents for a multi-step job with no manual orchestration from me.
How does a multi-agent workflow work?
One coordinator agent owns the job. It splits the work into sections and spins up sub-agents that run at the same time. For a newsletter:
- Research sub-agents find the news, events and listings
- Writing sub-agents draft each section
- An assembly sub-agent puts the issue together
The old design was one script doing the whole job, start to finish. With parallel sub-agents per section, I found builds both faster and less error-prone.
How good are the results?
Testing Claude Opus dynamic workflows, I automated a local newsletter to roughly 97% quality in about an hour. I expected a trade newsletter, like one for the insurance industry, to be even easier to automate than a hyperlocal one. More on the newsletter side in can AI write a local newsletter.
What else can multi-agent setups do?
- Sales follow-up. After a call, an agent reads the transcript, builds a proposal site, creates a payment link and sends onboarding once paid: automating the sales process.
- Writer and editor. I set up a writer agent and an editor agent in a shared workspace. The writer writes an article and tags the editor, and the editor comes to review it. I call it multiplayer AI: getting Claude, Codex and ChatGPT out of being a thing one person uses.
- Plan and redline. On big builds, one model writes the plan, a second model redlines it, and cheaper models do the searching and building: plan with one AI, build with another.
What goes wrong with multiple agents?
- Agent-to-agent loops. I forbid agents from DMing each other. They get into loops and burn thousands of tokens in an hour. Agents can message approved humans only.
- Too many pings. More agents means more reports. If you over-report to yourself, you stop paying attention completely. See AI agents in Slack.
- Cost. Use the top model for planning and a cheaper one for implementation. Images and video are the main token burners.
- The last 10%. Complicated multi-step workflows are still where AI needs the most human iteration.
Answered by TJ Larkin. From partner calls, product tests and team calls in 2026. Last updated .