I was twenty-four when a man twice my age handed me a folder and said: “If this network stops, trains stop.”
I laughed.
He didn’t.
That was the last time I found the sentence funny. Everything I’ve learned about surviving a career happened in the thirty years after it.
Here’s the part nobody warns you about: I have outlived every technology I was hired for. The equipment I was trained on is scrap. The protocols I memorised are footnotes. The vendors I built my early reputation around have been merged, rebranded, or quietly discontinued.
And I’m still here.
Not because I’m brilliant. I’ve watched brilliant people get walked out of buildings.
Because of one rule I didn’t understand until year fifteen.
The rule you’re expecting
You already know what you think I’m going to say.
Keep learning. Reinvent yourself. Stay current.
Every career article ends there. Learn the new thing. Get the certification. Adapt or die.
It isn’t wrong.
It’s just not the rule.
I know engineers who did all of it. Certified in everything. First in the room on every new platform. Genuinely current, genuinely skilled.
Some of them are consultants now. Not by choice.
Because “stay current” is a treadmill with no finish line, and there is always someone younger running it faster and cheaper than you. If your only asset is knowing the newest thing, your value resets to zero every four years and you compete, forever, against people with more energy and fewer mortgages.
That’s not a career.
That’s a subscription you keep paying.
The rule that actually held
Here it is.
Stay close to the systems that cannot be switched off.
That’s it. That’s the whole thing.
Not the newest system. Not the most exciting one. Not the one with the best conference talks and the nicest documentation.
The one where consequence lives.
The one where somebody, somewhere, cannot afford for it to stop.
Trains. Water. Power. Emergency networks. Payment rails. Hospital systems. The unglamorous infrastructure that carries real weight and gets discussed only when it fails.
Everything about my career that worked came from proximity to consequence. Everything that stalled came from drifting away from it.
Why consequence compounds and novelty doesn’t
Think about what actually happens when a system cannot be switched off.
Nobody rips it out. They layer on top of it.
Which means the knowledge of how it was built, why it was built that way, and what will break if you touch the wrong thing — that knowledge doesn’t expire.
It compounds.
I have sat in rooms where a decision worth millions turned on one person remembering why a link had been configured a certain way in 2009. Not a certification. A memory. A piece of context that existed in one head and nowhere else.
That person is not replaceable by someone younger and cheaper.
That person is not replaceable by a model, either, and I say that as someone who works with these tools every day. The tool can tell you what the configuration says. It cannot tell you which of the three people who signed off on it was lying about the timeline.
Novelty knowledge depreciates. Consequence knowledge accrues interest.
Most people spend their careers buying the depreciating asset.
What this looks like in practice, and what it costs
I want to be honest about the price, because most career advice sells you the upside and hides the bill.
Staying close to consequence is not fun.
It means the four-hour migration window at three in the morning, because that’s the only time the system can be touched. It means the maintenance contract handover nobody scheduled, where the outgoing vendor has stopped caring and the incoming one doesn’t have the passwords yet. It means 1,613 devices past end-of-support that everyone has agreed to not talk about, and being the one who counts them anyway.
It means your LinkedIn will never look as exciting as the person doing generative AI pilots at a startup.
It also means that when the reorganisation comes — and it always comes — the conversation about your role happens differently.
I’ve been through more restructurings than I can reconstruct from memory. Every single time, the same pattern: the roles that got cut fastest were the ones furthest from consequence. Strategy functions with no operational surface. Innovation teams with impressive decks and no system anyone depended on.
The people who ran the thing that couldn’t stop were never in the first three conversations.
Sometimes we were in the fifth. But by then the panic had passed, and panic is what makes bad decisions about people.
The trap inside the rule
Now the part that took me another decade to learn, because the rule has a failure mode and I walked straight into it.
Proximity to consequence makes you safe. It does not make you visible.
I spent years believing the work would speak. That if the network held, someone upstairs would know why it held.
They don’t. They can’t. A system that never fails produces no evidence of the effort keeping it up. That’s the cruel arithmetic of infrastructure: your best work is indistinguishable from nothing happening.
So the rule has a second half, and without it the first half traps you.
Stay close to consequence. Then translate it upward.
Learn to say, in language a CFO understands, what the deferral actually costs. Not “these devices are end-of-support.” Instead: here is the failure probability, here is what four hours of downtime costs this business, here is the curve of what waiting another year does to both numbers.
The day I learned to write that sentence, my career changed more than it had in the previous ten years of technical work.
Not because the technical work stopped mattering. Because it finally became legible to the people making decisions about it.
Most engineers never make this jump. They resent having to. I resented it too. I thought translation was politics, and politics was for people who couldn’t do the real work.
I was wrong. Translation is the work. The system doesn’t just need to hold — someone has to fund it holding, and that decision is made in a room where nobody speaks your language.
If you’re twenty years behind me
Three things.
Ask where the consequence sits. In your company, in your sector, find the system that cannot be switched off. It may not be the one with the budget or the attention. Go there anyway.
Stop optimising for interesting. The most valuable position I ever held was, on paper, the least exciting one available. Boring compounds. Excitement resets.
Build the sentence. Whatever domain you’re in, learn to state its risk in money and time to someone who will never understand its mechanics. If you can’t, you will spend your career being overruled by people who can.
Thirty years in, that’s what I have.
Not a technology. Not a certification. A habit of standing close to things that matter and being able to explain, in plain numbers, what it costs when nobody does.
The trains still run.
Most days, nobody notices.
That’s the job.