From Chad Wallace

Forged in the grind. Built for the next operator in it.

Twenty-two years on a single industry through-line. Two years building production AI on Claude's API. BecomingIQ is the floor. The platform is FlowState IQ. The install practice is Elevate IQ. All three grew out of operations work that asked for them.

DirectChad Wallace· BecomingIQchad@flowstateiq.com

Twenty-two years.One industry.

OPERATOR-BUILT AIFloor · BecomingIQ
Platform · FlowState IQ
Practice · Elevate IQ
Run on Claude's API · Built from inside the work
§ 01Industry through-line

Twenty-two years, one industry.

I started on the wholesale floor at Town & Country Landscape Supplyin 2004, learning the industry by watching it move — what crews wanted on the truck, why some yards ran tight and others didn't, how the same product became three different conversations depending on who walked in. Then 10 years at Homer Industries, where I learned production at scale alongside selling it. Then SiteOne — sales at scale. Lavin Companies — building the operational infrastructure. Semper Fi — building the intelligence layer; Equipment, Procurement, Assets — value through new lenses.

Wholesale. Manufacturing. Distribution. Construction. Same industry, multiple positions inside it.

What that's given me is operational seeing— the discipline of reading the work the way an operator reads it, by living inside it. The thing I'm offering now grew out of that.

§ 02The work · operator-built AI

What I do now.

What I do now is build AI that fits operator-led businesses — software that holds the tacit knowledge an owner-operator carries in their head, written so the operation runs through it instead of around it. I'm a Builder. The work I take on targets the pain points that return value and time back to the operator who brings me in. I call it operator-built AI — built from inside the work, not delivered onto it.

The methodology came out of the same operations job that produced it: an anchoring question I keep returning to — what must be true for this to work?— that surfaces gaps when assumptions stop holding, paired with WHAT→HOW→WHO sequencing and a nine-stage build discipline I'd been practicing unconsciously for two decades before I named it.

BecomingIQ is the floor — where the proofs live before they become products, where the operator-built thinking is authored in the open. FlowState IQ is the platform the proofs graduate into. Elevate IQ is the install practice that brings them to the operator. Three altitudes. Same operator. Same discipline.

§ 03What's been built

What's been built.

The platform — FlowState IQ— runs on Claude's API. Five-layer prompt architecture governing 19 AI modes. A 22-agent ecosystem with two fabrication factories that produce new agents and skills through structured methodology. A 55-table production database holding the operator's data and the operator's rules. The discipline matters because the data does.

What landed in the platform wasn't the plan I started with. The methodology was discovered, not designed— it emerged from the operations work that demanded it, not from a blueprint I brought to the work. That's why it fits operators: it was produced by one.

Early engagements are in motion in the Green Industry I operate within — and the proofs live here at BecomingIQ before they cross over. Field Notes carries the running record of what's in motion, what I'm seeing, and what I'm thinking about.

§ 04The craft, listed

The craft, listed.

A short version of what I work with day-to-day.

AI systems on Claude's API
Multi-layer prompt architecture · 19 conversational and batch modes · 22-agent ecosystem · structured fabrication of new agents and skills · pattern libraries that accumulate across builds.
Methodology
What must be true for this to work? (the anchoring question) · WHAT→HOW→WHO sequencing · Builder Framework (9-stage, dependency-sequenced) · ground-truth verification · agent self-examination as a build discipline.
Platform engineering
TypeScript · Next.js · React · Supabase/PostgreSQL with row-level security · Python (FastAPI · MCP servers) · Vercel deployment · progressive migration discipline.
Data discipline
Schema design that doesn't break · classification pipelines that produce auditable structured output · ontology modeling across workflows, gaps, ownership, and systems.
Operator systems
QuickBooks · Microsoft platforms (SharePoint, Power Automate, Power Apps) · vertical-SaaS substrates already running.
§ 05Posture · what's anchored, what's open

What's anchored, what's open.

What's anchored

The platform is real and shipping. The discipline behind it has held across multiple build cycles, multiple agents, multiple shipped surfaces. The industry through-line is twenty-two years deep.

What's still open

Every operator who lands at this practice is part of the first round. The next install is the next stress-test of the practice's repeatability — the first time the methodology meets a shape the current ones didn't predict. I'd rather name that openly than pretend the practice has more anchor-points than it does.

§ 06Engagements · trust earned, then extended

Trust earned, then extended.

I'm a Builder. First engagements are designed together — targeting the pain points that quickly return value and time back to the operator who brings me in. The work produces quality outputs that earn the trust before any second engagement is presented.

Tier 1 · Earn the trust

Designed together. Scoped tight. Shipped fast.

A defined, testable install against pain points named at intake — by the operator, with me. Quality outputs measured against the work that was producing the friction. The build either earns its keep or it doesn't — that's the gate.

Tier 2 · Once earned

The same foundation, extended.

The operator-built foundation extends — additional surfaces, deeper installs, the practice running across the operation instead of inside one corner of it. Same discipline. Same data. More of the operation legible to itself.

No commitment to Tier 2 at signing. Tier 1 has to earn it.

If you're an operator inside the band I'm describing — landscape, supply, manufacturing, contractor-adjacent, $500K to $25M — and what's here lands, I'd want to hear it. If it doesn't land, I'd want to hear that more. The shape of the read sharpens fastest when an operator inside the band tells me where it doesn't quite fit.