0x01 / IDENT AI & AGENT SPECIALIST · MARKETING TEAMS

I put AI to work inside marketing teams.

TIME▼ COST▼ OUTPUT▲ AGENT LAYER · ONLINE

Agents, automations and custom tools that take over real processes — reporting, creative ops, media buying, content, CRM — so they run in a fraction of the time, at a fraction of the cost, with better output. Built by an operator who managed $35M+ across 20 markets before automating any of it.

SEE WHAT I AUTOMATE ▸ [email protected]

▸ performance teams · brand teams · content teams — if it's repeatable, it's automatable

PROCESS MAP · LIVE nodes 0 · edges 0
▸ idle hover a process to query it · drag to rearrange · click a hub to jump

0x02 / READOUT OPERATOR RECORD — THE PART AI CAN'T FAKE

0+
years inside marketing teams
node/ops.tenure
$0M+
media managed hands-on
node/ops.spend
0
markets operated
node/ops.geo
0
marketers led day-to-day
node/ops.team
0+
AI systems in production
node/sys.live

0x03 / PROCESSES WHAT I AUTOMATE

Pick a process. Any process.

PRC/01 · DATA

Reporting & Dashboards

Weekly and monthly reports compiled, annotated and delivered by an agent — no pulls, no pivots, no Sunday-night decks.

MANUAL → AGENT-RUN
PRC/02 · CREATIVE

Creative Ops

One brief in — on-brand variations and ad copy at scale out, with creative fatigue flagged before performance drops.

ONE BRIEF → FULL SET
PRC/03 · MEDIA

Media Buying Ops

Launch checklists, placement hygiene, bleeding checks and budget shifts — verified and executed by agents, not at 11pm.

CHECKLISTS → GUARDRAILS
PRC/04 · CONTENT

Content & Social

One asset repurposed across every channel; calendars drafted, scheduled and kept on voice without the copy-paste grind.

ONE ASSET → EVERY CHANNEL
PRC/05 · CRM

CRM & Lifecycle

Segments, flows and win-back triggers built from actual behavior — lifecycle marketing that reacts in hours, not quarters.

GUT FEEL → BEHAVIOR
PRC/06 · QA

Compliance & QA

Every landing page scanned against policy rules before spend touches it — risk scored, fixes drafted automatically.

SPOT CHECKS → EVERY PAGE
PRC/07 · INTEL

Research & Intel

Competitor moves, market shifts and platform changes monitored continuously and briefed to your team on schedule.

AD-HOC → ALWAYS-ON
PRC/08 · OPS

Internal Ops

SOPs, onboarding docs and meeting follow-ups that write themselves — institutional knowledge that stops living in heads.

IN HEADS → IN SYSTEMS

Not on the list? Doesn't matter. If your team does it more than once a week, it's a candidate — performance, brand, content or ops.

0x04 / PROTOCOL HOW AN ENGAGEMENT RUNS

From time sink to agent-run, in five steps.

01MAP

Sit inside your workflow

A week in your actual operation — listing every repeatable process, where the hours go, and what good output looks like.

02SCORE

Rank by payback

Frequency × time × error cost. We pick the two or three processes where automation pays for itself fastest — no AI for AI's sake.

03BUILD

Build on your stack

Agents, automations or custom tools — wired into what you already use: ad platforms, Sheets, Slack, CRM, your data. No rip-and-replace.

04DEPLOY

Prove it in shadow mode

The agent runs next to the manual process until its output is verifiably better and cheaper. Then — and only then — we switch.

05TRANSFER

Hand over the keys

Documentation, training and guardrails so your team owns and extends the system. No black boxes, no dependency on me.

0x05 / BRIDGE WHY ME

I ran marketing before I automated it.
That order is the moat.

● OPERATOR — LIVE EDGE

Ran the machine

Owned spend, targets and a team. Lived inside the constraints automation has to respect — approval chains, policy walls, creative fatigue, messy data, real deadlines.

$35M+ spend20 markets15-person team
● BUILDER — AUTOMATION EDGE

Builds the machine

Encodes that operating knowledge into agents, memory and workflows. The systems I ship know how the work is actually done — because I did it.

agent meshesmemory systemscustom tools

0x06 / PROOF SYSTEMS ALREADY RUNNING

Shipped, not slideware.

SYS/01 · openclawPRIVATE

OpenClaw

A private multi-agent command center — one console commanding a mesh of agents across marketing work, with persistent memory and task routing.

agent meshmemory vaultroutingtelegram ops
SYS/02 · lp-complianceLIVE

LP Compliance Scanner

Chrome MV3 extension that scans a landing page against 20 policy rules, outputs a weighted risk score, and uses Claude to draft the fixes.

20-rule engineweighted scoreAI-written fixesMV3
SYS/03 · learning-appSHIPPED

AI Learning App

A mobile learning product with voice-driven interaction — speech transcribed and routed into adaptive, model-generated lessons.

FlutterFirebaseOpenAIWhisper
PROMPT DEMOS→PRODUCTION AGENTS MORE HEADCOUNT→MORE LEVERAGE HOURS OF GRUNT WORK→MINUTES OF REVIEW

0x07 / ENGAGE WAYS TO WORK TOGETHER

Four ways in.

Fixed-scope or outcome-tied — priced around the process, not the hours.

ENG/012–4 WEEKS

AI Process Sprint

Map your processes, ship the first one or two automations, and prove the time saved — fast, contained, low-risk.

ENG/02FIXED SCOPE

Agent Build

One painful process in, one production agent out — built on your stack, deployed in shadow mode, documented and handed over.

ENG/03ONGOING

Embedded / Fractional AI Lead

Your part-time Head of AI. Roadmap, prioritization, builds and team enablement — compounding leverage, quarter after quarter.

ENG/04TRAINING

Team Enablement

Hands-on training built from real workflows — your marketers running and extending their own agents within a quarter.