On Legacy

On Legacy

November 15, 2025·
robcost

On my last day at AWS, after 8 and a half years, I posted one final message to my team. It was that meme from The Usual Suspects, you know the one, with the caption “and like that, he’s gone.” Then I removed myself from the Slack channel and shut down my laptop.

I’d left companies before. Twice from Microsoft, and both times I’d felt that surge of panic, the “oh shit, what have I just done?” moment. But this time, closing that laptop after nearly nine years, I felt… relief. I was ready. I was tired. And more importantly, I was moving on to build something that mattered to me in a different way.

What I didn’t expect was how much I’d think about what I was leaving behind. Not the projects or the processes, those were never mine, not really. But the people, the conversations, the small moments where I might have helped someone see their situation more clearly. That’s the stuff that sticks with you when you leave. And it’s the stuff that makes you wonder: what actually counts as legacy anyway?

The False Legacy

Here’s what big tech wants you to think legacy looks like: patents with your name on them, new services you helped launch, processes you created that outlive your tenure. Visible artifacts. Things that can go on a promotion packet or a LinkedIn profile.

And look, there’s nothing wrong with those things. They matter to the business. They might even matter to your career. But there was always this implicit pressure at AWS, like at most big tech companies, to pursue those visible legacies if you wanted to advance. If you wanted the promotions, the pay bumps, the recognition, you needed to be visible.

That was never me. I didn’t want that visibility. I just wanted to help.

When I joined the UN team in 2020, and then moved to Geneva in 2022, I knew from the start that I was transitioning into an “elder” role of sorts. I’d been at AWS for four years in Australia by then, long enough to understand how to survive and thrive in an organisation that size. I knew part of my role, maybe the most important part, was going to be guiding and growing others professionally and personally. Not because it would get me promoted, but because others had done it for me, and I needed to pay it forward.

The thing is, that kind of legacy doesn’t show up in quarterly reviews. There’s no metric for “number of colleagues who left your team but took your advice with them.” No dashboard tracking “conversations that helped someone understand they weren’t being singled out, they were just caught in organisational dynamics they couldn’t see clearly.”

But that’s the legacy I actually left behind at AWS. Not the global GenAI enablement program I started with the UN, though yes, that’s still running, morphed into “GenAI + Data” from what I see on LinkedIn. Not the technical strategies or the customer wins. The thing I hope made a difference were the conversations I had with peers as they navigated their way through AWS. The stories I told about my own mistakes. The advice I gave that helped them see their situation with a bit more clarity.

That knowledge might still be passing through the organisation. Someone I helped might be helping someone else now. That’s legacy, even if my name isn’t attached to it anymore.

The Conversation That Matters

There was one teammate in Geneva who sticks with me. He was having a hard time with leadership, really struggling. And when you’re in that situation, everything feels personal. Everything feels like an attack. You can’t see past the immediate pain to understand the organisational dynamics at play.

We had a series of conversations. Not formal mentoring sessions or scheduled one-on-ones, just… conversations. We reasoned through the situation together. Took the emotion out of it, or at least tried to. Empathised with the positions of others, understood the drivers that made them act the way they did. None of it was malicious. Most of it wasn’t even about him. It was just organisational machinery grinding away, and he happened to be caught in the gears.

I think I helped him come to peace with his future, which unfortunately turned out to be outside AWS. But he moved on successfully. And I hope, I really hope, he took that framework with him. Not just the specific advice, but the way of thinking through difficult situations. The ability to see past the emotional reaction to the structural reality underneath.

That’s what I mean by legacy. Not what you leave behind in the organisation, but what you leave behind in people. Knowledge and wisdom they can pass on to others. A way of seeing that helps them help someone else down the line.

The Geography of Leaving

Geneva and Switzerland are intertwined with my AWS exit in a way that makes them hard to separate. Leaving the company meant leaving the city, leaving the country, moving the family back to Australia. That time in my career and that time in my life are wound together so tightly that thinking about one inevitably pulls up the other.

We spent nearly three years there. We traveled Europe extensively as a family, made lasting memories, expanded our worldview in ways that wouldn’t have happened if we’d stayed in Australia. My kids got to see a different way of living. My wife and I got to experience what it meant to work and build a life in a completely different context.

When I think about what I left behind in Geneva now, it’s not with regret. It’s with gratitude. We always knew we were coming back to Australia eventually, which meant we never had that anxious feeling of “we’ll never experience this again.” We were living on borrowed time, and we knew it, which somehow made the whole experience more vivid.

Now that we’re back, and now that I know we won’t move back to Europe, I think about that life differently than I did when we were living it. Not as something lost, but as something I got to have. The difference is subtle but real.

There’s something about leaving a place that’s different from leaving a role. When you leave a company, the organisational memory might hold onto you for a while, but eventually you fade. When you leave a place, a city, a country, you leave behind a version of yourself that only existed there. The person I was in Geneva, doing that work, living that life, doesn’t exist anywhere else. That version of me is still there, in some sense, even though I’m not.

Maybe that’s true of all the places we leave behind. The versions of ourselves we lived into in those contexts remain there, even as we move on to become someone else somewhere else.

The Different Kind of Legacy

I’m building myEdi.ai now, an AI tutoring platform for kids aged 8-15. And my relationship with legacy has shifted in ways I didn’t expect.

When I’m writing code, or documentation, or support articles, any kind of artifact for the company, I do think about handoff. I need to make sure someone else could pick this up and see it as quality content. That’s the builder’s instinct: create things that can outlive your involvement with them.

But that’s not the legacy that matters most anymore.

What matters now is the impact the product, and eventually the company, can have on my target customers. Have I helped kids learn better? Have I reduced stress on parents? Have I helped these families and kids have better life experiences? That’s the legacy I want to ensure is meaningful. Not the working process or the elegant tool, but the lives that are different because this thing exists.

Here’s the interesting tension: at AWS, people knew me personally. They came to me for advice. They knew Rob helped them. There was direct recognition, direct connection. With myEdi.ai, the kids and parents who use it will never know it was me personally. They’ll never know my name or my story. The product will help them, but I’ll be invisible behind it.

And that’s not just okay, it’s actually freeing. The work matters more than the credit. Impact at scale matters more than individual recognition.

I can still mentor people one-on-one. That’s not something I can only do inside a company. I’ll still have those direct connections, those individual conversations that might help someone see their path more clearly. But now I’m also building something that scales beyond what I could do in person. Something that can help thousands or tens of thousands of people I’ll never meet.

That’s a different kind of legacy. Not better or worse, just… different. And more aligned with where I am now, at this stage of my career and life.

What You Actually Leave Behind

I declined a management role at AWS two years before I left. I knew at the time that decision was closing doors. I knew it meant my path was limited and my time would eventually come to an end. But I wanted to be in control of the timing, not have the timing forced on me by organisational changes or strategic pivots three layers above my pay grade.

So I spent those last two years helping level up the people around me. Sharing what I’d learned. Paying forward what others had given me. And when the timing was right, when I was ready, when my family was ready, I started the exit planning. Six months of careful transition, documentation, knowledge transfer.

The team I helped build is still there, roughly the same size, still doing the enablement work I created and orchestrated. I haven’t heard much from them since I left, which is actually a good sign. It means things kept going without me. I’d like to think I did a good job showing them how to be trusted advisors, not just tech sales support people. How to engage at the strategic level, talking about long-term planning and solving long-standing issues, not just putting out fires.

But I don’t know for certain. I don’t have visibility into what’s working and what isn’t. I don’t know if the things I thought were important turned out to be important, or if the team has evolved in directions I never anticipated.

And that’s fine. That’s how it should be. They don’t need me anymore. That’s the whole point of mentoring people well and building processes that can run without you. You make yourself obsolete, and if you’ve done it right, that obsolescence feels like success, not failure.

The Gratitude Loop

Here’s what I’ve learned about legacy after 25+ years in tech, across Microsoft, Dell, AWS, and now building my own thing:

Legacy isn’t what you leave behind. It’s what you give forward.

It’s not the patents or the products or the processes that survive after you’re gone. It’s the conversations you had that helped someone see their situation more clearly. It’s the advice you gave that they took with them to their next role, their next company, their next challenge. It’s the perspective you shared that helped them understand they weren’t being singled out, they were just navigating complex organisational dynamics that are easier to see from the outside.

And now, building myEdi.ai, it’s the impact your product has on people you’ll never meet. The stress you reduce, the learning you improve, the family experiences you make better by solving a problem that matters.

You can’t control whether any of this actually lasts. You can’t control whether the team keeps doing things the way you set them up, or whether they evolve beyond recognition. You can’t control whether the people you mentored remember your advice or pass it on to others. You can’t even control whether your startup succeeds or fails, no matter how hard you work or how good your product is.

What you can control is whether you approach each role, each conversation, each piece of work with the intention to help. To leave things a little better than you found them. To give the people you work with the tools and perspectives they need to navigate their own paths.

Everything else is impermanence. And impermanence isn’t something to fight, it’s something to accept with gratitude.

I’m grateful for the time in Geneva. Grateful for the team I helped build, even though they don’t need me anymore. Grateful for the conversations I had with colleagues who were struggling, even though I don’t know what became of most of them. Grateful for the eight and a half years at AWS, even though I left with relief, not regret.

And I’m grateful now for the chance to build something that might help people at scale, even if they never know my name.

That’s the thing about legacy. The best kind doesn’t have your name on it at all. It’s just… there, helping, long after you’ve moved on to the next thing.

And like that, you’re gone. But what you gave forward keeps going.

And like that