Executive AI Consulting · Sufficiently Advanced AI

AI is already remaking your company.
Are you the one deciding how?

It spread faster than anyone architected it, and now you're accountable for what it becomes. From here it goes one of two ways: a slightly cheaper version of the company you already have, or capabilities the business has never had. Only one of those compounds.

See how it works Book a 30-minute call
Claude Certified Architect
Most advisors stop at strategy. I build the system. Over a decade in enterprise architecture and AI. Claude Certified Architect
The Real Choice

You can't cut your way to a future.

The tools are already in the building. McKinsey's latest survey of nearly 2,000 companies found 88 percent using AI somewhere — and the high performers among them were nearly three times more likely to have fundamentally redesigned the work, not just added tools to it. PwC asked 4,454 CEOs what AI had returned; 56 percent said neither revenue nor savings.

Adoption is common. Redesign is rare. And aiming AI at "today's work, fewer people" keeps you on the wrong side of that line — your people hear that goal exactly as it sounds, and no one runs toward it.

Optimize the old thing

Same work, fewer people. A one-time bump, then a plateau, and a team quietly bracing for the next cut.

Build something new

Capabilities the business never had. Work redesigned around what's now possible. People who see a place in what's coming and pull adoption forward instead of resisting it.

Does This Sound Familiar?
"I keep wondering whether this is actually working, or whether we're just busier and spending more. And whether I'd even know the difference."
The question I hear most often from executives

Decisions get made, then drift

Someone decides something, it cascades, and a later conversation quietly goes a different way. No one connects the dots, and the work looks green right up until it's badly off track.

Everyone's using it; nobody's steering it

One person built a workflow in Claude. Another team bought a different tool. A third pastes into whatever's open. It spread on its own, and no one's pointing it anywhere.

The tools run; the company hasn't changed

You've got accurate summaries and copilots deployed. But you're running the same business, slightly faster. Nothing you couldn't do a year ago.

Who does this work

Scott Jennings

Founder · Sufficiently Advanced AI · Hillsborough, NC

Scott Jennings
LinkedIn GitHub Email Substack
Claude Certified Architect badge
Claude Certified
Architect

For a decade I architected enterprise software — stood up the architecture team at a global EdTech SaaS, then led systems and data through a company-wide transformation. In every role the hardest problem was the same, and it was never the technology.

Context sits in silos. Decisions don't reach the people they affect. The mistakes that cost the most are the ones no one catches in the moment.

I'm a Claude Certified Architect, and I built an open-source AI system end to end (that's IMI) to surface exactly that. I stay hands-on from design through implementation, and bring your people through the change as much as the architecture.

The Approach

I learn how you actually work — then build what advice alone can't.

Most AI projects fail because they run on your documentation — how things are supposed to work, not how they do. The real logic lives in decisions, and decisions happen in conversation, not the wiki. The conversation is the source of truth.

The Accelerator

I start from a working blueprint, not a blank page.

IMI is my reference build for AI-native organizations — the architecture and operating point of view I bring to every engagement, then adapt to the stack you already run. You don't adopt my system; I use it to get yours right, faster.

See IMI
How We Work Together

Start with an assessment. Continue as your architect.

You don't need it figured out to start. Most engagements open with a short, fixed-scope assessment you own outright. Go further or don't. From there, I stay on as the architect who turns it into working AI.

Where we start · The assessment · Fixed fee · Fixed scope · 3–4 weeks
What I do
  • — Find where AI is already in use — official tools, personal accounts, and the workflows nobody formally approved
  • — Interview the people doing the work, not the people who documented it
  • — Measure two or three real workflows end to end — every handoff, every re-keyed field, with the hours attached
  • — Trace where the reasons behind your recurring decisions actually live — in a record, or in someone's head
What you get
  • An inventory of every AI tool in use: what it costs, who depends on it, and what would break if it disappeared
  • A measured baseline for those workflows — because "it feels faster" won't survive budget review
  • A decision map: which of your important recurring decisions are written down, and which walk out the door with the person who holds them
  • A prioritized roadmap, and a first build scoped concretely enough to hand to anyone — including a hire
What your people stop fighting · Illustrative
The same problem, solved twice

Two teams spend a quarter building the same fix because neither can see the other's work. Clear it, and that effort comes back as capacity for the work only they can do.

The change that didn't reach everyone

A decision shifts, scope changes, a project gets refocused, but the update doesn't reach everyone, and work keeps moving on the old understanding, sometimes without the one team whose perspective would unlock it. Close the gap and the work reflects the current call, not last month's.

The judgment that should travel further

Your best people are sought out for their judgment, but it only reaches the rooms they're in. Give it reach, and it guides far more of the work without asking more of them.

Book a 30-minute call
Is This Right for You

Complex enough to feel it. Empowered enough to fix it fast.

Size isn't the signal; an unowned operating layer is. Maybe it lives in a founder's head at a 20-person firm, maybe it sprawled across teams at a 300-person one. What matters is an operator who feels it and can act — before your AI choices harden into infrastructure you're stuck with.

You're already spending on AI

Claude or other tools are in real use, whether that's you and a few power users or a dozen teams. The investment exists; the architecture doesn't. Nothing has to change to start.

No one owns how it fits together

Maybe the whole operating picture lives in your head. Maybe adoption ran ahead of anyone architecting it. Either way, no one holds the top-down view, and when someone leaves, their piece goes with them.

You can decide without a committee

Whoever sees the problem can act on it. Weeks to start, not quarters. No seven-vendor RFP for a four-week assessment.

Not the right fit?

If the goal is the same work with fewer people, it's not a fit. Plenty of firms do cost takeout. This is for leaders who want to enable their teams to build what the business couldn't before.

Common questions

Will my people embrace this, or resist it?

It depends on what they think it's for. People resist AI when the message is "fewer of you" — no amount of training fixes a destination that sounds like a layoff. When the goal is better tools and less sprawl, buy-in gets far easier. I spend time with your team early, and build work that amplifies what they do, not watches them.

We're hiring an AI lead. Do we still need this?

Especially then. Left alone, a new AI leader spends their first quarter discovering what already exists — the tools in use, the workflows, where the data lives, who can say yes to it. The assessment is that discovery done before day one, so your hire starts with an inheritance instead of an investigation. If the role isn't filled yet, the ongoing partnership covers the gap: the mandate gets defined, the baseline gets measured, and the first workflow reaches production while you search — so you hire against evidence, not a guess.

We've already invested in AI. How is this different?

Most companies have tools: copilots, a strategy deck, a few automations. The real question is whether any of it builds something the business couldn't do before — and whether your people use it or quietly work around it. I start from how you actually operate and architect from there, so what gets built fits your business, not a generic rollout.

How does this work, and what am I committing to?

It starts small: a short, fixed-scope assessment you own outright, whether or not you go further. From there it becomes an ongoing partnership — strategy, architecture, and build, effectively a fractional Chief AI Officer at the altitude of your leadership team. You work directly with me throughout: a senior engagement scoped to the outcome, not a seat or a tool.

Executive Briefs

The diagnostics I run, written down.

Short briefs for executives working through an AI ownership role — what to put in place, how to interview for it, and how to tell whether the work has actually changed. Direct download, no signup.

What You Can’t Rent

Seven in ten companies hiring an AI leader hire no one else. What an internal owner can hold, what only an outside architect supplies, and which one you are missing.

Read the brief

Five Questions for an AI-Enabled Role

A twenty-minute check on one job posting: did the work change with the tools, or were the tools just added to the work?

Read the brief

Interviewing for Expertise You Do Not Have

How to evaluate candidates who know more about AI than anyone on your side of the table — without pretending to be a technical grader.

Read the brief
All briefs
Start Here

Let's see if I can help.

Thirty minutes, no charge, nothing to prepare. Tell me where you are with AI — what's running, what's working, what you're not sure about — and I'll tell you where I'd look first.

Book a 30-minute call

Or write first, and I'll reply within one business day.

✓

We'll be in touch.

Expect a reply within one business day. We'll schedule a short conversation to figure out if there's a fit.