I've built, bought, sold and listed several companies in my career.
Raised capital.
Made bets.
Lived with the consequences.
But sitting on the board of Retraite Québec several years ago was something else entirely.
You’re not allocating venture capital.
You’re allocating retirement.
Millions of people.
Decades of work.
A level of responsibility that doesn’t leave room for narratives or hype.
What struck me the most wasn’t the scale. It was the discipline.
Every technology investment had to answer a different set of questions:
Not “is this exciting?”
Not “is this innovative?”
But:
Does it actually work?
Will it still work under pressure?
Do the people behind it truly understand what they built?
And if something goes wrong… can they explain it?
You start to see things differently.
You stop optimizing for upside alone.
You start optimizing for resilience.
Because failure isn’t just a missed quarter.
It’s public trust.
And it changes how you look at software. Especially today.
There’s more code being shipped than ever. Faster than ever.
Often without the same level of understanding behind it.
That might work in a startup environment.
It doesn’t translate the same way when public money is involved.
That experience stayed with me.
It still shapes how I build today.
What I pay attention to.
What I question.
What I’m not willing to compromise on.
Because at some point, every system ends up being relied on by someone who doesn’t have the option for it to fail.
And when that moment comes… “it works” isn’t enough.