Skip to content

Who this is for

The audiences that recognise these moments, each with the ones it recognises. Membership is declared by every moment and never listed here.

  1. Solo builder 14 One person, several assistants, and no one else to ask what was decided yesterday.A few sessions a day across one or two checkouts, architecture held in one head, no review step and no handover to anybody but tomorrow.
  2. AI-native development team 25 Several people and more assistants than people, working the same repository at the same time.Agent sessions of several vendors in parallel worktrees and branches, work handed between humans and workers, and no single place that says who holds what.
  3. Engineering lead 28 Accountable for what shipped, without having watched any of it being written.Sets outcomes, reviews results, answers for delivery; reads state rather than transcripts, and has to trust what the state says.
  4. Platform and developer productivity team 20 Owns the rules, the tooling and the execution environment that everybody else works inside.Writes conventions once and needs them to hold across many repositories, many providers and many workers who never read the wiki.
  5. Open-source maintainer 13 Reviews contributions from people and machines they have never met, against conventions only they know.Asynchronous, unpaid, interrupt-driven; the review queue is the product and every contribution arrives without the context that would make it correct.
  6. Agency or consultancy 16 Many clients, many repositories, and context that must transfer between people in hours rather than weeks.People move between codebases constantly; every move is an onboarding, and every onboarding is paid for by somebody.
  7. Enterprise engineering 22 Answerable to auditors, not only to users: what was done, by what, under which policy, with what evidence.Standards are mandatory, changes are traceable, and "the assistant did it" is not an acceptable provenance record.
  8. Research and experimental engineering 10 Many more experiments than results, and the expensive part is remembering which ones were already tried.Short-lived branches, disposable sessions, several models and providers compared, and findings that matter long after the session that produced them ended.

The number on each card is the moments that name it, counted from the moments themselves. An entry's own file lists none.