Slide 01

You Cannot Change the Organization. You Can Route Around It.

CTO + VP Engineering + Board

Core claim

Every organization has a Terry right now — a leader who got the AI transformation mandate and has no real authority to transform anything.

Terry got promoted. They are leading the AI transformation. Meanwhile a four-person startup just raised $30M and signed Meridian Financial — Terry's first customer, the logo that got the company through the IPO. The startup processes the same volume with 12 engineers that Terry's organization handles with 200. The Meridian CTO's quote: "We needed a partner who moves at AI speed, not committee speed." Terry's company is an "agile" company. Three transformations in ten years. Seven full-time agile coaches. Still takes six weeks to ship a button.

The mandate vs. the authority Terry was given a mandate to lead AI transformation. Terry was not given authority over the VP of Engineering's process, the CPO's roadmap commitments, or the General Counsel's approval workflows. These are not the same thing.

Slide 02

12 Engineers. 200 Engineers. Same Volume. You Are Running Out of Time.

Market signal

The startup 12 eng

Four Stanford kids. $30M Series A. Processing the same volume that takes Terry's organization 200 engineers to handle. Already signed Meridian Financial — Terry's first customer and IPO logo.

Terry's org 200 eng

Nine years of transformation. SAFe, Spotify squads, OKRs. Seven full-time agile coaches. Release committee meets every other Thursday. Six weeks to ship a button when nothing goes wrong.

The gap 16.7x

Sixteen point seven times as many engineers to process the same volume. That is not a productivity difference. That is a business model difference. And Meridian Financial just voted on which model they prefer.

"We needed a partner who moves at AI speed, not committee speed."

Meridian Financial CTO, on why they left for a 12-person startup

Slide 03

Sarah Closes in Hours What Takes Others Days. Nobody Knows. She Did Not Tell Anyone.

Finding the movers

What Terry found in the Git history

Nine people shipping faster than they should be able to. Seven of them using AI. Nobody talking about it. All underground.

Sarah has been using AI tools for three months. She pays for them herself. The last time she suggested changing how things work, she spent six weeks in meetings with the transformation team who wanted to turn it into a pilot program with success metrics and a steering committee. She decided it was easier to just do it quietly.

What this means The innovation you want is already happening. It is hiding from you — because your organization made hiding cheaper than sharing. That is a leadership problem, not a technology problem.

What Terry did next

Terry did not announce a pilot program. Did not form a committee. Did not ask Legal to draft an AI use policy in isolation.

Terry spent all of Friday with procurement and legal. Got enterprise licenses approved. Got legal sign-off on an acceptable use policy. Ran it through security review. Four hours. Then told the nine people they had found: "This is legal. It's approved. You're not going to get fired for using it. The only question is what we're going to build with it."

The unlock That changed the whole conversation. Engineers who had been sitting on techniques that could save everyone else weeks — opened up. Because the risk was gone.

Slide 04

80 Hours Up Front to Save 240 Hours Per Team Per Quarter. No Theoretical Examples Allowed.

The evidence model

What a Community of Practice produces

Regular ceremonies. A charter. A steering committee to govern the community.
One agile coach asked if they should build one. Terry said no.
The last Community of Practice generated fewer contributions in two years than the repo generated in two months.
Theoretical examples. Best practices documents nobody reads. Aspirational guides that do not connect to production systems.

What the internal GitHub repo produced

All 30 prompts they tried before finding one that worked. Why the first approach failed.
The guardrails they built. The things they still do not trust AI to do.
Actual time investment: 80 hours up front to save 240 hours per team per quarter.
Rule: you have to show me the production system you built with it. No theoretical examples.

Slide 05

300 People in the Channel. Posts Every Day. No Mandate. No Celebration. Just Evidence.

Culture change without permission

The #shipped-with-ai channel

No celebrations. No transformation theater. Just: what you shipped, how long it took, link to the PR, and one thing that surprised you.

Three weeks after launch: 300 people in the channel. Posts every single day. Engineers getting competitive. A frontend engineer saw the backend team eliminating API boilerplate. Asked if it would work for Redux actions. Two days later, she shipped it.

"That's viral," Terry's friend said. "Yeah. And I didn't mandate any of it. The culture is changing and I'm not even trying to change it."

The mechanism Evidence spreads laterally without anyone issuing a directive. The people who see results want to replicate them. The channel is not a celebration — it is a proof of concept library that updates itself.

Test environment maintenance — eliminated

The first visible win: completely automated test environment maintenance. Something everyone hated. Something visible enough that other teams would notice.

Terry did not announce it. They just stopped having the problem. Took three days before another team asked how they did it.

The lesson Pick something everyone hates. Fix it. Do not announce it. Wait for someone to ask. The credibility of the question "how did you do that" is worth more than any transformation kickoff.

Slide 06

Terry Cannot Change the Organization. Terry Changed What the Organization Believes Is Possible.

The constraint reality

The honest realization

"I can't actually change the organization. Like, at all. VP of Engineering built the entire process. CPO controls the roadmap with commitments from two years ago. General Counsel controls Legal."

Terry knew this going in. But internalized it over time. There is no direct authority to change the release committee, restructure the team topology, or override the CPO's roadmap. The organizational gravity is enormous and distributed.

So the play is not to change the structure. It is to make the structure irrelevant in the places where it can be made irrelevant — by showing what is possible within it, at the edges, with the people who are willing.

What you can change

You can find the people who are already moving. You can remove the risks keeping them quiet. You can build the evidence repository. You can start the channel. You can pick the visible problem that everyone hates and solve it quietly.

When 300 people are posting daily evidence of what is possible — when the test environment maintenance problem simply goes away — the people who control the structure start asking different questions. Not "should we transform?" but "how did they do that and how do we get more of it?"

The shift From: "how do we get people to adopt AI?" To: "how do we get more of what Terry's group is already doing?" That is the shift. It happens through evidence, not mandate.

Slide 07

Are You Waiting for Permission to Transform, or Are You Building What You Can Within What You Have?

Decision close

The question for every Terry

You have the mandate. You may not have the authority. You definitely have the Git history. Start there.

The 12-engineer startup that took your customer is not waiting for the committee to finish its charter. They ship while you are still deciding what to measure. The gap compounds every quarter. Two years from now it will be insurmountable for most organizations that are still in "planning the transformation" mode.

Terry's insight: you do not need to change everything to start building something real. You need to find the movers, remove the risks, build on evidence, and let the results make the argument for scale. That is a strategy you can execute without the authority you were not given.