How an engagement runs.

Six phases, from the first call to 90 days after handoff. Every engagement I run follows this structure, though the work inside each phase changes.

  1. phase_01

    intro & fit

    1 call · 30 min

    A structured, free conversation. I ask about your organization, what you want agents to do, what's in the way, and your timeline. By the end we both know whether there's a fit and which service matches your problem.

    deliverable: Fit memo (one page, my notes, shared with you within 24h)
    $ ls -la phase_1/
    intro_call_notes.mdfit_memo.pdf
  2. phase_02

    scoping

    1–2 weeks · async + 1 call

    If we move forward, I review what you've built, where you sit on the agent stack, and what's running in production, and I run my full readiness assessment. You see how I think before you decide whether to keep going.

    deliverable: Scoped engagement proposal: milestones, deliverables, capped T&M or fixed price
    $ ls -la phase_2/
    readiness_assessment.mdscope_proposal.pdf
  3. phase_03

    kickoff

    1 day · on-site or remote

    The engagement officially starts here, with the full team. We map the stack, the people, the risks, and the first wins worth chasing, and write a 90-day plan with named roles.

    deliverable: Kickoff deck + 90-day plan + roles + risk register
    $ ls -la phase_3/
    kickoff_deck.pdf90_day_plan.mdrisk_register.csv
  4. phase_04

    build · pair · advise

    varies · 2–10 weeks

    The work itself: embedded pairing for a build sprint, weekly working sessions for advisory, and written check-ins every two weeks either way. I work inside your Slack and your repos.

    deliverable: Whatever the milestones specified, tested, documented, and owned by your team.
    $ ls -la phase_4/
    pull_requests/framework_docs/weekly_notes/
  5. phase_05

    handoff

    1 week

    I leave documentation your team can extend, and run a working session with the engineering and ops people who will keep the system running. I record it so future hires can onboard from it.

    deliverable: Handoff doc + recorded walkthrough + 30-day check-in scheduled
    $ ls -la phase_5/
    handoff.mdwalkthrough.mp4next_steps.md
  6. phase_06

    follow-through

    30/60/90 days

    I check in at 30, 60, and 90 days, unbilled, to see whether the work held up. Most clients continue into an ongoing relationship when it makes sense for them.

    deliverable: Check-in calls + optional retainer conversion
    $ ls -la phase_6/
    30day_review.md60day_review.md90day_review.md

The phase_01 intro call is free and takes thirty minutes, with no deck.

$ book the intro call
> esc
$ keyboard_shortcuts.md esc

// navigation

  • ⌘K or ctrl+K — open command palette
  • ? — toggle this overlay
  • esc — close any overlay
  • / — jump to command palette input

// command palette

  • ↑ ↓ — navigate results
  • ↵ — select / navigate
  • type a tag like #agents — search by tag
  • type a service like workshop — find related pages

// 404 page

  • cd <path> — navigate
  • ls — list pages
  • help — show commands
  • clear — clear history

// agent swarm

  • click anywhere — spawn a task
  • drag a rectangle — dispatch a team
  • hover over a worker — see its model
based on conventions from cmd+k, vercel, linear, and most claude code workflows