Business: Technology · Lesson N.tec.8

The technology function's OS: the library that works for you

A common course delivers a lesson; a strong course delivers infrastructure. This lesson joins the module's choreographies into a living system, your technology function's OS, which improves itself with every good delivery.

Examples for

Look at your screen right now. There's a prompt saved in a notes app that always works to triage a new ticket. There's a postmortem document you duplicate every incident and tweak by hand. There's that script you put together to sweep for technical debt and almost never run because you forgot where you saved it. Each piece works on its own. The problem is they're scattered, spread out, dependent on your memory. This lesson is the moment to bring it all together in one place, with a name and an order, and turn that pile of good hacks into a system that works for you.

Let me tell you something about courses. A common course delivers a lesson: you watch, do the exercise, close the tab, and three weeks later you remember roughly the concept. Think with me: what stayed in your workday? Almost nothing. A strong course is something else. A strong course delivers infrastructure, something that stays running after you closed the tab, that changes how you operate Monday morning. This entire module was built to leave you with infrastructure, not with a memory. This lesson is where we install it for good.

The core idea of this lesson. Your technology function's OS is a living library with four shelves: the prompts that work, the delivery templates in the canonical format, the agents and flows that run on their own, and the audit checklist every delivery passes through. The criterion for what becomes what is simple: a repeatable, stable task becomes a template or an agent; a task that changes every time stays more in your hands, with AI helping. And the trick is that this system improves itself: every good delivery you make becomes a new piece in the library. You're not leaving here knowing about AI in technology. You're leaving with AI installed in the way your team works, and nobody takes that away.

01The difference between a lesson and infrastructure

Let's call it what it is. What separates whoever watches an AI course and keeps doing the same from whoever watches it and moves up a level isn't the amount of memorized prompts. It's whether that turned into a system or turned into a note.

A note is fragile. It depends on you remembering, on you finding the file, on you having the energy to rebuild the prompt in a Friday-at-5-PM rush with an open incident. A system is the opposite: it's ready, has a fixed place, opens fast, and works even when you're tired or when it's someone else on the team on call. The economic question behind this is direct. What's an hour of engineering worth? Every time someone on the team rebuilds from scratch something that's already been done ten times, the team is paying that hour for not having organized. The OS is what stops charging that bill.

LOOSE PIECES prompt runbook script depends on the memory of whoever's on call SYSTEM prompts templates agents and flows fixed place, opens fast, survives an on-call switch

The difference isn't magic, it's arrangement with intent. And that's exactly what we're going to do now. Fair?

02The OS's four shelves

The technology function's OS isn't a product you buy. It's a four-shelf structure you build with what you already produced in this module. Each shelf has a clear function.

prompts that work delivery templates agents and flows audit checklist the function's day-to-day work faster and audited the four shelves feed every delivery the team makes

Notice this isn't theory. You've already produced a piece for each of these shelves throughout the module. The OS is the act of taking them out of the drawer and putting them on the right shelf.

03The criterion: what becomes a template, what becomes an agent, what stays manual

The question that trips people up most here is: automate what? The answer has a single criterion and it fits in one sentence. The more repeatable and stable the task, the higher up the automation scale it climbs. The more it changes every time, the more it stays in your hands, with AI just helping.

changes every time always identical stays manual becomes a template becomes an agent the more stable, the higher automation climbs

This criterion spares the team two expensive mistakes. The first is automating what changes, and getting stuck with an agent that gets it wrong outside the script in an incident nobody foresaw. The second is leaving in your hands what's identical every time, and continuing to pay engineering hours out of reluctance to build the flow. You want each task at the right height on the scale. Fair?

Learn more: why "build versus buy" almost always stays manual

It's worth a concrete example of why an important task doesn't always climb the automation scale. The decision to build a tool internally or buy from a vendor seems repeatable, because the team faces this choice several times a year. But the content of each decision changes: the vendor is different, the internal maintenance cost is different, the dependency risk is different. Turning this into an agent would be dangerous, because a fixed criterion doesn't capture each case's nuance. What it safely becomes is a template: a matrix with the same columns every time (total cost, lock-in risk, integration time, support), filled in fresh for each decision. AI helps fill it out fast; reading the matrix stays yours, because it's judgment, not repetition.

04The trick: the system that improves itself

Here's the part that turns the OS from a dead file into something alive. A well-built OS doesn't stay still. It grows with every use.

Here's how it works. The team closes an incident this week, say a postmortem that came out particularly well written, with a clear root cause and a specific prevention action. In the old way, that work dies at delivery: someone posts it in the Slack channel and life goes on. In the OS, it doesn't die. The prompt that helped write that postmortem becomes a piece on the prompts shelf. The structure that worked becomes or reinforces the template. The check that caught an important detail becomes one more line in the audit checklist. Every good delivery leaves a sediment in the system.

The compound effect of this is big. In month one, the OS has the basics. In month six, it has the entire library of the best ways to do each thing, distilled from dozens of incidents, PRs, and real decisions. The team gets faster not because the agent got smarter, but because the team's system became more theirs. The practical rule is one: every good delivery ends with a question, what's worth keeping from this? That question is what keeps the OS alive.

And notice that's the opposite of starting from zero. Most teams start every incident or every decision at square one, wrestling with the same prompt again. Whoever has an OS starts from the accumulated. That's the advantage that shows up slowly and then becomes impossible to catch up to.

05The module's choreographies already go into the OS

Now it's time to close the loop. Everything you practiced in this module wasn't a loose exercise. Every choreography is already a piece ready to go on the shelf. To recap:

If you want to see where the technology function's OS fits in the bigger picture of the journey, it's your instance of what the orchestration track calls infrastructure: each piece of it is a packaged capability the team reuses instead of reinventing. Technology was just the domain where you built the first one. The method is the same for any area of the business.

And that's why the message from the start of this module holds true here: you're not leaving here knowing about AI in technology. You're leaving with AI installed in the way your team works. The difference is huge: knowing fades, systems stay. You didn't finish a module, you built infrastructure. And nobody takes that away.

Do it now

Do it yourself

Open a blank document and give it the title: Technology Function OS, your real task. Create the four shelves as sections:

  1. Prompts that work. List three to five prompts you or the team tested in this module and that delivered good results (e.g., "new ticket triage", "incident log correlation", "deploy summary for the team"). Give each a descriptive name and paste the prompt.
  2. Delivery templates. List the canonical formats you already have or want to have: the postmortem, the build versus buy matrix, the technical debt map. For each, write the section structure in one line.
  3. Agents and flows. List what already runs or should run on its own: backlog triage, the debt sweep. Mark what already exists and what's still to be built.
  4. Audit checklist. Write the five checks every agent PR passes before becoming a delivery (does the diff match the intent, is the blast radius what you'd expect, and so on).

At the end, classify each item on shelf 3 by the lesson's criterion: is it repeatable and identical (becomes an agent), repeatable with new content (becomes a template), or does it change every time (stays manual). This document is the index of your team's OS. From today on, every good delivery ends with the question: what's worth keeping from this?

Practice

1. What's the criterion for deciding what becomes an agent, what becomes a template, and what stays manual in technology?

2. What makes the technology function's OS a living system, not a dead repository of scripts?

3. What's the difference between a module that delivers a lesson and one that delivers infrastructure, in this track's sense?

For the board

On what remainsknowledge fades, systems stay. The team walks away with running infrastructure, not notes.
On the criterionstable and repeatable becomes an agent. Repeatable with new content becomes a template. What changes every time stays in your hands.
On compoundingevery good delivery leaves a sediment, and the system becomes more the team's with each round.
What did you think of this page?
Would you recommend this page to someone on your team?