Raw Content

How I Run Three AI Models in Parallel Without Losing My Mind

**By Katie Parrott

Every**

Core Argument

Katie Parrott describes a fundamental shift in how knowledge workers operate: rather than choosing between being a maker (requiring deep focus) or a manager (coordinating others), AI has created a hybrid role she calls “model manager”—someone who strategically allocates attention across multiple AI processes running simultaneously.

Key Framework: The Allocation Economy

Parrott references her own concept of the “allocation economy,” where value derives from deploying attention strategically rather than diving deeply into single tasks. She notes: “The old paradigm assumed you were either building or coordinating—never both at once. AI breaks that assumption.”

This challenges Paul Graham’s 2009 “Maker’s Schedule, Manager’s Schedule” framework, which treated these roles as incompatible. AI enables a different working pattern entirely.

Real-World Workflow

On productive days, Parrott:

Sends ChatGPT a transcript to extract quotes and structure
Uses voice dictation (Monologue) for revisions while waiting
Has Claude analyze published pieces for style patterns
Cycles between tasks as models process

The result: “My actual writing time has dropped 40 percent, but my weekly output has tripled.”

Managing Context Switching Costs

Despite gains, context switching creates significant overhead. Parrott’s four practical boundaries:

Compartmentalize rigorously — Separate chats per project preserve context
Leave breadcrumbs — One-line notes about next steps prevent lost threads
Limit active tasks — Never exceed three simultaneous processes
Capture immediately — Export finished work to permanent storage before continuing

The Paradox

This workflow enables solo operation at small-team scale but raises questions about sustainability and whether increased output justifies the cognitive load.