Author frames AI agent workflow as 'agent hopping' to distinguish it from multitasking
A writer describes managing multiple AI agents as a distinct workflow that avoids the cognitive costs of traditional task switching.
Key points
- The author spent five hours managing three projects simultaneously using AI agents.
- Fiona Fung of Anthropic describes running multiple agents while navigating meetings.
- The author distinguishes this workflow from multitasking by citing Vibe Coding concepts.
A technology writer reported spending five hours at a Munich coworking space managing three separate projects with the assistance of AI agents, describing the experience as distinct from traditional multitasking.reported The author noted feeling mentally exhausted after the session, comparing the sensation to fatigue from back-to-back interviews.A writer found that using AI to manage multiple projects felt different from traditional multitasking because they only switched tasks when agents finished their work.The author posits that AI-mediated task switching occurs at natural seams between work packages, potentially bypassing the cognitive costs associated with interrupting deep focus.
Evidence: reported
Why working with AI agents is so tiring (and what to do about it) – Tech, work, and what comes nextaifleet-lobsters
What was released or claimed
In a blog post titled Why working with AI agents is so tiring, the author detailed a workflow involving an open source contribution, a blog analytics redesign, and a brain dump that became the post itself. The author stated that AI agents did the heavy lifting, allowing them to review results, make decisions, and write the next set of instructions.
Who it affects
The material specifically mentions Fiona Fung, who manages the Claude Code and Cowork teams at Anthropic. She described on Lenny’s Podcast how she kicks off multiple agents simultaneously and knits through meetings while they work. This suggests the workflow is relevant to technical leaders and developers.
What is still unknown
The author admits uncertainty about whether "agent hopping" is an established term or if the described workflow truly avoids the performance degradation associated with task switching. The post questions whether decades of research on multitasking are wrong or if this is a fundamentally different activity.
The takeaway
The author argues that working with AI agents resembles running a professional kitchen, where the human acts as a head chef directing line cooks. This metaphor implies that switching occurs at natural hand-offs between tasks rather than interrupting deep focus, allowing the user to choose the moment of transition.
What changes for users
Users may need to reframe their understanding of productivity when using AI agents. According to the source material, Instead of viewing simultaneous project management as harmful multitasking, the author suggests treating it as agent hopping, where the user reviews outputs and provides new instructions at specific seams between tasks.reported This approach relies on the user maintaining control over the timing of switches, potentially reducing cognitive load compared to traditional context switching.
Why it matters
- Data science teams can adopt this "head chef" model to manage multiple parallel workflows without the cognitive penalty of traditional multitasking.
- The distinction between agent hopping and multitasking helps teams justify using AI for parallel task execution rather than sequential processing.
- Understanding this workflow allows teams to structure their prompts and review cycles around natural hand-offs, improving efficiency.
- The reported fatigue suggests that even with AI assistance, managing multiple streams requires mental energy that teams must account for in scheduling.
Topics: AI agents, productivity, Anthropic, workflow, multitasking
Updated
Then / now / next
Multitasking was widely considered a productivity anti-pattern that degraded performance and focus.
Some users describe managing AI agents as a distinct workflow that avoids interrupting deep thought.
Teams should observe whether this workflow reduces fatigue or creates new cognitive burdens over time.
Story timeline
- Then Multitasking was widely considered a productivity anti-pattern that degraded performance and focus.
- 05 Nov 2026 Why working with AI agents is so tiring (and what to do about it) – Tech, work, and what comes next
- Now Some users describe managing AI agents as a distinct workflow that avoids interrupting deep thought.
- Next Teams should observe whether this workflow reduces fatigue or creates new cognitive burdens over time.
The strongest doubt
Counterpoint from the sources
Psychologists have repeatedly measured that task switching degrades performance far beyond the time taken for the switch itself, raising questions about whether agent hopping truly avoids these costs.
Ask the article
Local keyword search of this article; no generated answers.Did this article give you the evidence you needed?
Your choice stays on this device. No public tally.Questions readers ask
What is agent hopping?
Agent hopping is a term used by the author to describe switching between AI-managed projects at natural hand-off points, rather than interrupting deep work.
Who else uses this workflow?
Fiona Fung of Anthropic described kicking off multiple agents at once and knitting through meetings while they work.
Does this replace multitasking?
The author argues it is not multitasking because the user chooses the moment of switching and does not interrupt their own thought process.
Connected stories
Reading receipt
- The author spent five hours managing three projects simultaneously using AI agents.
- Fiona Fung of Anthropic describes running multiple agents while navigating meetings.
- The author distinguishes this workflow from multitasking by citing Vibe Coding concepts.
