First 90 days — operator packet (East Tennessee two-site)¶
What this is¶
A bounded, field-facing runbook for the first 12 weeks after T0—pilot execution only. It does not replace the narrative arc on First 90 days — Phase 0–1; it packages what the operator actually does each week, what evidence must exist, and when to stop.
Pilot justification (explicit): Pilot work is justified when T0 is set on the Decision memo, the family stays inside Approved now + Approved only for pilot, and EG rows are pursued at pilot bar—not production / G1 posture (Execution gates — Phase 0–1). Do not treat this packet as permission for Forbidden rows on the memo.
Aligns with: narrowed Phase 0–1 scope · EG gates · Field verification program — Phase 0–1 · Operator burden review (hostile) · decision memo.
Out of scope here: months 4–24 detail, Phase 2 CAPEX, herd scale, salary narrative—use First 12 months after day 90.
Weekly priorities (weeks 1–12)¶
Rule: Each week, pick at most one “hard” slice (long drive + heavy field) unless surge—Operator burden review.
| Week | Priority theme | Ship this week (max 3 outcomes) |
|---|---|---|
| 1 | T0 + books + stop rules | T0 date on memo; hour log week 1 started; vision stop-rule template copied locally |
| 2 | Legal surface + map authority | V8 started (ticket/letter path); one canonical map layer name; trip purpose codes started |
| 3 | Labor truth | V11 week 3 logged; coverage matrix holes named (no heroics) |
| 4 | Checkpoint A | Decision checkpoints; adjust week 5–8 if overload |
| 5 | Land intel (cheap) | V1 desktop + first Demory walk tied to field program §1; V3 photos started |
| 6 | Water + access | V2 question list moving; water points on map (I1); bad-weather access log once |
| 7 | Markets paper | V5 first sale bills; V6 first processor call log row |
| 8 | Checkpoint B | EG-LEV pilot pass honest? NS-2 triage budget real? Trim pilots if not |
| 9 | Insurance + haul | V9 quotes collected; B3 haul bracket documented as range, not fantasy |
| 10 | Connectivity (pilot) | V10 pin screenshots started; WAN placement/obstruction log per first-90-days connectivity table |
| 11 | G8 + comms stability | Manual water (or chosen G8 asset) log in progress or complete; no second sensor class |
| 12 | Checkpoint C + handoff | Day-90 “done enough” items from First 90 days § End of day 90 checklist; schedule first 12 months review |
Field verification tasks (mapped)¶
Execute rows from Field verification program in this order of emphasis—not all sections every trip.
| Week band | Demory (SITE_FARM) emphasis |
Claxton (SITE_HOME) emphasis |
|---|---|---|
| 1–4 | §1 orienting + boundary/fence walk starter; §2 access lane photo | §4–5 utility / ISP / OSSS truth path started (desktop + any field) |
| 5–8 | §3 drainage / sediment eyes after rain if possible; §1 fence segments deepen | Same + §6 gateway location paper only unless pilot already approved |
| 9–12 | §4–5 power / pump spot checks without new always-on RF; §7 sensor mount sites scout only | §6 homelab / WAN power intent for CPE |
Cadence: Follow field program § “Bundling”—every Demory trip bundles access + one infra slice.
Infrastructure setup tasks (90-day cap)¶
| Task | Site | Done when | Gate / stop |
|---|---|---|---|
| Primary route bad-weather log | Demory | One logged attempt or documented deferral with date | Supports EG-SITE-2 pilot |
| Water points on map | Demory | All known points pinned + Unknown list | V2 / I1 |
| CPE / WAN power intent | Claxton | Ticket or two-person shutdown story on paper | Connectivity table Days 31–60 row |
| Spot fence / escape fix | Demory | Only safety per memo pilot infra row—not perimeter MVP | Memo Forbidden guard |
Sensor / network / platform tasks (90-day cap)¶
| Task | Done when | Stop |
|---|---|---|
| Declare stack scope | EG-PLT-0: written “no k3s” or “k3s pilot = {nodes}” | If undeclared, freeze cluster feature creep |
| G8 manual baseline | Asset chosen; paper/spreadsheet log started week ≤8; complete by T0+120d if not by day 90 | No sensor install without baseline path |
| One RF / one gateway pilot | Matches DR-5 class scope; no second farm WAN / HaLow while DR-1 open | CS-5 |
| V10 cell at pins | Notebook/screenshots started week ≤10 | No “telemetry drives trips” claim without V10 track |
| Triage time budget | H_TRIAGE_BUDGET_WK set in ROI worksheet or family equivalent | NS-2 |
Documentation updates required¶
| Artifact | Owner | By |
|---|---|---|
Missing data register rows |
Ops lead | Weekly status + owner (EG-LEV-1) |
Local evidence package tables |
Ops lead | After each field visit bundle |
| Hour log (V11) | Named family | Weekly through week ≥8 inside 90d window |
| WAN / reliability notebook | Tech lead | First entry by day 30; weekly by day 90 |
| Pilot one-pager (one sensor class, kill threshold) | Tech lead | Before any install (EG-SNS-2) |
Stop conditions (non-exhaustive)¶
Freeze or cut the week’s plan if any apply—then document and reschedule:
Decision memoForbidden row would be violated by the next action.- EG pilot fail in a domain you are about to spend in (e.g. register orphan rows before claiming “ready for contractor talk”).
Automation stop rulesNS-1–NS-3, CS-1, CS-4, CS-5 after one remediation attempt.Operator burden reviewcollision week: Tech lead and Field lead same person and both had surge—drop one pilot thread.- R17 signal: triage + field + off-grid same week with no coverage—invoke coverage matrix or cancel non-welfare work.
Evidence collection requirements¶
| By | Minimum evidence | Feeds |
|---|---|---|
| Day 30 | V8 path started; V11 ≥4 weeks; map authority; one Demory field visit log (field program); WAN obstruction/placement log started | EG-SITE-1 pilot, EG-LEV-2 pilot |
| Day 60 | V1 + V3 usable for conversation (not perfect); V5/V6 first rows; V10 notebook started | V tasks; first-90-days days 31–60 |
| Day 90 | First-90-days end checklist satisfied or dated slip with which item carries to T0+120d | Handoff to First 12 months |
Decision review checkpoints¶
| When | Review questions (family) | Outcome |
|---|---|---|
| End week 4 | Still inside memo Approved / pilot only? Any Forbidden pressure? EG-LEV pilot rows owned? | Adjust weeks 5–8 scope |
| End week 8 | V11 on track for 8 weeks total? NS-2 triage budget real? Exactly one telemetry pilot thread? | Kill or defer sensor/WAN extras |
| End week 12 | First-90-days end checklist met or slips dated? G8 done or scheduled by T0+120d? No G1 / production composite claimed without evidence? | Written plan for days 91–180; handoff to First 12 months |