Wireframes - AI integrations (W23–W29, W35)

Every AI surface follows the same four rules: show the grounding, expose confidence, require human approval before customer contact, and capture accept/edit/reject feedback that flows to the model console (W28). All generation runs on the local purpose-tuned LLM (3.3) unless a task is explicitly switched to a hosted API through the adapter interface. v4.3 adds the model lifecycle pipeline (W29) - how an adapter graduates from eval to production with a human signing the promotion - and interaction review (W35), which turns each call, email, and visit into playbook-fit coaching and tracked commitments. Deliberately not built: a general-purpose chatbot (every surface here is task-shaped) and a standalone lead-scoring screen (the score lives on W19; its evaluation lives on W28).

W23 - AI composer (in-context drafting)

Drafting from inside a job or Contact 360: template intents, tone/length controls, personality match applied, and a grounding panel showing exactly which retrieved facts (pgvector) the draft used, plus consent and pricing-floor checks. The AI never sends - approve/edit/reject is the human gate, and that feedback trains the next adapter (W28).

Draft email - Dana Harmon local LLM · adapter v12
Follow-up Financing info Schedule ask
Tone Factual Warm Length Short Detailed Regenerate

Subject: Your underlayment question + itemized financing breakdown

Hi Dana,

Following up on your question about the synthetic underlayment - attached is the spec sheet, including the valley ice-and-water detail we discussed on the quote. Since you were comparing monthly options, the Better option at $18,750 works out to $228/mo on the 84-month plan - the full amortization table is attached, no hard credit pull to see terms.

No rush on a decision. Happy to answer anything else line by line.

Jordan · Summit Roofing

Approve & open in editor Edit Reject AI never sends - a person always does · feedback trains adapter v13

Grounded on

Quote Q&A: underlayment vs. felt question (Jul 14)
Website: /financing ×3, calculator hotspot (heatmap)
Quote #1042: Better · $18,750
Playbook: C-Analytical - data first, no pressure

pgvector retrieval · sources clickable

Checks

Email consent: drip-eligible

Claims match quote data

No pricing below floor

W24 - AI nodes in the automation canvas

Extends the W4 canvas with AI node types - generate message, classify reply, extract fields - under guardrails: a confidence branch routes anything below threshold to a human task, an approval-gate node is required before customer-facing sends, and each AI node has a model selector exposing the local-adapter vs. hosted-API choice from 3.3.

Automation: reply handler Live
Run history + Node

TRIGGER

Reply received

SMS or email

AI · CLASSIFY

Intent + opt-out

local · adapter v12

BRANCH

Confidence ≥ 80%?

else → human task

AI · GENERATE

Reply draft

personality-matched

GATE · REQUIRED

Human approval

before any send

ACTION

Send reply

Branch: opt-out → suppression list (no gate needed - always immediate) Branch: <80% → task for rep

Node config - AI · Classify

Model

Local · Qwen2.5 + adapter v12
Hosted API (via adapter I/F)

Labels

interested · scheduling · question · not-now · opt-out

Confidence threshold

80% · below → human task

W25 - Photo damage detection review queue

Vision AI over the media hub (W20): proposed hail-hit annotations with per-detection confidence, accept/adjust/reject controls, and a certification path into the insurance packet. The principle: AI proposes, a named inspector certifies - nothing reaches an adjuster without human sign-off, and the packet records who approved what. Accept rates feed the model console (W28).

Damage review - Olsen hail claim 14 detections · 3 need review
Certify & add to insurance packet
hail hit · 96% hail hit · 93% review · 61% review · 54%
IMG_2044 · N slope · confident needs review
Accept detection Adjust box Not damage selected: review · 61%

Claim summary (draft)

Confirmed hits9
Pending review3
Rejected2
Slopes affectedN, W
Storm matchJun 12 · 1.5"

Certification

Inspector J. Diaz certifies each detection. AI output is never sent to an adjuster without human sign-off; the packet records who approved what.

Model

vision v4 · accept rate 87% · feedback → model console

W26 - Inbox triage & reply classification

Inbound texts/emails auto-classified (interested · scheduling · question · not-now · opt-out) feeding the automation engine. The compliance hook is the point: AI-detected opt-outs write to the suppression layer immediately with no approval gate, while low-confidence classifications route to a human queue. Override rates feed W28.

Inbox - 23 new AI triage on
All Needs review (3) Interested (6)
Wu, K. - "Yes let's do Thursday morning if the crew…" scheduling · 97% → booked via automation
Pratt, S. - "STOP texting me" opt-out · 99% → suppressed · all channels
Harmon, D. - "Does the 84-month plan have a prepayment…" question · 94% → draft ready (W23)
Unknown # - "my brother said you did his roof, we're at the lake house…" 62% - human queue
Bell, R. - "Not this year, check back in spring" not-now · 91% → nurture Mar 2027

Opt-outs today: 2

Written to suppression instantly - no approval gate on opt-outs, ever. Synced to CRM + data engine.

Last 30 days

Auto-handled78%
Human queue22%
Overrides11% ↑

W27 - Ask-your-data analytics

Natural-language questions over the warehouse, answered with the chart plus provenance - which continuous aggregate, which filters, the generated SQL - so answers are auditable, not oracular. Read-only access against Timescale aggregates; no AI write path to data.

Ask your data local LLM · read-only
cost per signed job for the June hail mail campaign, by week Ask

Cost per signed job - Jun hail mail

Overall: $201 per signed job ($3,410 / 17 signed)

$412
Wk of Jun 16
$247
Jun 23
$168
Jun 30
$121
Jul 7

Falls over time as mail-piece-2 responses convert · suggested follow-up: "compare to Meta ads campaign"

Provenance

Source: campaign_outcomes_weekly (continuous aggregate)

Filter: campaign_id = jun-hail-mail

Measure: spend ÷ jobs_signed

Attribution: tracking # + QR, 90-day window

View generated SQL →

W28 - Model ops & eval console (admin)

The governance screen the v4 LLM decision demands: active base model and LoRA adapter version, held-out rubric scores side-by-side against the hosted-API baseline (the comparison the adapter escape hatch enables), GPU cost vs. estimated API spend, per-surface human feedback rates (W23/W25/W26), drift alerts, and the retune queue with rollback targets.

Model ops - AI layer Serving healthy
Eval history Queue retune (v13)

Base model

Qwen2.5-14B

Apache 2.0 · adapter v12 (LoRA)

Serving

vLLM · 2× L4

p95 1.9s · 99.7% uptime

GPU cost / mo

$1,140

vs est. hosted API $3,400

Escape hatch

Adapter I/F

hosted API swappable per task

Eval rubric - local v12 vs hosted baseline (held-out set, blind-scored)

Playbook accuracy
8.8 /9.1 Draft usability
9.0 /8.9 Reply classification F1
.94 /.92 Open-ended reasoning
7.1 /9.0

▮ local · | hosted baseline - open-ended gap accepted: no product surface depends on it (tasks are narrow & tuned)

Human feedback (30d)

Composer accept / edit / reject64 / 29 / 7%
Damage detection accept87%
Inbox triage overrides11% ↑

Triage override drift > 10% - retune suggested

Adapter registry

v12 · activeJun 30
v11 · rollback targetJun 2
v13 · training queuepending

W29 - Model lifecycle & promotion pipeline

How a model change ships: offline eval against the rubric → shadow mode beside the live adapter (output never shown to users) → canary A/B through the experiments console (W30) → promotion behind a required human sign-off → post-promote monitoring with a warm rollback target. No stage can be skipped, including for hosted-API swaps through the adapter interface; accepted/edited/rejected outputs from W23/W25/W26 become the next version's training set.

Model lifecycle - adapter v13 (candidate) stage 2 of 5 · shadow
Training data diff Rollback serving to v11

1 · OFFLINE EVAL

Passed

Rubric ≥ v12 on all four axes · triage F1 .96 (+.02) · held-out set, blind-scored

2 · SHADOW · RUNNING

Day 9 of 14

Runs beside v12 on live traffic, output never shown · agreement 94% · disagreements sampled for review (31 open)

3 · CANARY A/B

Queued

10% of composer traffic via Experiments (W30) · metric: accept rate · auto-stop on reject spike

4 · PROMOTION · GATE

Human sign-off required

Eval + canary results attached · sign-off: ops lead · promotion is logged, attributable, reversible

5 · MONITOR

Post-promote

Accept/override drift watched (W28) · one-click rollback target kept warm (v11)

No stage can be skipped - including for hosted-API swaps via the adapter interface Feedback from W23/W25/W26 becomes v14's training set

W35 - Interaction review & relationship health

Every interaction - call, email, text, site visit - reviewed by the local LLM against the customer's personality playbook and stored history: playbook-fit checks, sentiment, and detected commitments tracked to completion (a promise made on a call becomes a task). Aggregates roll up to a relationship-health score on the customer side and a coaching view on the rep side. Guardrails are part of the design: per-state recording consent with played disclosure, reps see their own reviews first, manager views lead with aggregates, reviews coach and never auto-discipline, and disputes route to a human.

Interaction review - Dana Harmon local LLM · playbook C-Analytical
This customer Rep coaching view

Call · Jul 14 · 12m

fit 8.6 · positive

Email · Jul 14

fit 9.1 · sent via W23

Site visit · Jul 9

fit 7.2 · notes thin

Text · Jul 2

fit 8.8 · quick reply

Call · Jul 14 - AI review (vs. C-Analytical playbook + stored history)

✓ led with data ✓ no pressure close ✓ answered line-by-line △ install timeline left vague

Commitments detected

Send amortization tabledone · Jul 14 email
Check prepayment penalty with Wisetackopen · task created
Hold Jul 20–21 crew slotdone · calendar
Coaching note (private to rep + manager): analytical buyers want dates - give a concrete install window next touch Agree · Dispute

Relationship health

82 ▲ +6 this wk

responsiveness · sentiment · commitments kept · quote engagement

Guardrails

Call recording follows per-state consent; disclosure played on record.

Reps see their own reviews first; manager view is aggregate-led.

Reviews coach, never auto-discipline; disputes route to a human.

← Wireframes: Content & growthSlide deck →

AYO & Ryzo Platform - hypothetical product specification and wireframe set, v4.0 (July 2026). Feature inventory derived from public JobNimbus and SumoQuote materials; all screens, names, and figures are illustrative. Back to index