Matt Wolfe interviews Peter Yang about the small, reusable AI workflows behind his creator business, then builds simpler versions of some himself. Yang still works with video and thumbnail contractors and approves customer-facing changes.
Separate daily attention from long-term direction
- A twice-daily chief of staff returns linked, numbered mail and calendar priorities. Yang delegates drafts and scheduling by number but reviews outgoing email.
- His advisor reads a short plan of goals and values. Wolfe makes his own version and reserves edits to that document for himself.
- Daily task triage and a roughly monthly goals review have different rhythms. Neither requires a single agent with unrestricted access to every app.
Improve one workflow at a time
- Yang’s content skills handle guest research, interview preparation, video feedback and publishing drafts. A visual review app lets him leave time-specific notes for an editing agent.
- His skill editor looks across multiple conversations for repeated corrections. Wolfe selects individual changes rather than accepting an opaque pile of new rules.
- Wolfe demonstrates a bass-practice log and recurring check-in as a smaller closed loop: purpose, data source, useful output and delivery schedule.
- Caveat: the title’s “runs his entire business” overstates the demonstrated autonomy; people still choose titles, approve messages and edit long-form video. The same guest covers overlapping examples in a separate same-day tutorial.
“Pick one workflow that would actually save you time, give it good context … and keep improving it as you go.” — Matt Wolfe, 27:01