6-year anniversary reflection

The moment you finally get the thing you wanted, strangely, the doubt shows up. I graduated in May 2014 and have worked ever since — 12 years. On 5/26 I hit six years at Instagram and a career the 24-year-old me wanted. I thought I’d feel like I’d arrived. Instead I feel uneasy, and I’ve come to think the unease is the more useful thing to write down.

Six years taught me how to operate here — how to shape a decision before it’s made, read what people actually want, lead a team. But the same years taught me something I’m slower to admit: it turns out all that control over the work around me was mostly borrowed. Below are the lessons from those years — some from where I had a real sense of control, and some from where I had none at all.

1. Most influence happens before the meeting, not in it

My biggest lever was the prep nobody sees — the work that quietly decides a meeting before it starts. I hate being pushed into things, so I over-prepared just to avoid it. Oftentimes, whoever frames the problem before alignment starts usually lands where they wanted, while everyone else spends the meeting reacting. Everybody knows this in theory; less people do the boring prep. It’s the subtle influence without any real power.

2. People run on incentives, not orders

The most useful skill I picked up was understanding what a person actually wants, which predicts what they’ll do. It’s tempting to assume people just do what they’re told. In practice, you see how hard someone runs when a project serves their goals, and how little they give when it doesn’t.

So part of the job became simple diagnosis: what does this person want, would the proposal get them closer to it, and if not, what else can I put on the table. This pays off even with people you’ll never work with — once you can see someone’s incentives, their choices stop looking irrational and start making sense, and a lot of friction just disappears - I simply stop wasting time forcing alignments. What actually moves people is the overlap between what needs doing and what they already want.

3. You can’t make everyone succeed

I tried to make everyone on my team succeed, and slowly learned I could only control where I put my own energy. I came prepared: I’d map out people’s projects for them, estimate their impact, try to keep things balanced across the team. Part of it was the TL role, but mostly I just wanted everyone I led to find a path.

What I learned is that energy spent on people who aren’t reaching for it is energy taken from the ones who are — I was helping people who hadn’t asked, while the quietly ambitious got less of me. Now I save the deep, long-term attention for the people who actually care. It sounds cold, but it’s the only thing that scales.

3. With AI, the rare skill is judgment, not building

After a year of building with Claude Code almost daily, I’ve landed on this: AI is shifting the value of an engineering leader from building to judgment. Knowing how to build something, or how to break it down and hand it off, still matters — but it’s no longer the bottleneck.

When anyone can stand up a working version in an afternoon, the question stops being “can we build it” and becomes “should we, and at what cost.”

The trap I watch good engineers fall into, myself included, is rushing to build because we can, without asking why, or what we’re quietly taking on. The more valuable work now is the judgment around the building — what risk are we baking in, how would we measure it, and is this even worth building. That pulls engineers toward instincts that used to belong to other roles: a PM’s sense of what to build, a designer’s eye, a data scientist’s habit of measuring the outcome.

For lack of a better word, it adds up to taste. And I suspect taste — knowing what to build, and what not to build — is the next big growth area for engineering leadership. When anyone can build anything, the only real control left is deciding what’s worth it.

4. To lead is to go first

Leading, unlike managing, means going first — and you can only do that when you’re not afraid of the fall. Going first means trying the new thing, making the new mistake, taking the first round of criticism.

Two rules I’d stand behind: be willing to be the dumb person in the room, and lift people up wherever you can — that’s what makes them follow you without being asked. But I have to be honest: I lead “fearlessly” mostly because I’ve had air cover the whole time. Manager, director, on up — sponsorship for things I tried. Take that scaffolding away, and I don’t know how far I would have gone.

5. The voice you build is on loan

At year six, the catch is plain: almost everything I’ve built here rests on a platform I don’t own. Most of us are replaceable. The illusion that I’m important, I’m needed is easy enough to drop.

The stubborn one is the illusion of control — the belief that I steer my own work, when a few decisions far above me, or the broader economy, could change everything overnight without asking.

My title gives my words weight in a meeting; I’d lose that voice the moment I step off the platform, because it was never mine — it was on loan. That’s not a complaint. I’m holding a genuinely good hand. But a good hand is still cards someone else dealt, on a table someone else can fold.

The questions I’m left with are the ones I still can’t answer. Where should I be in a few years, when this run inevitably ends? What’s the end game I’m chasing — versus the one the platform chose for me? Who am I, and what is my voice worth, without the title?

A week ago, these questions stopped being abstract. On 5/20 the layoffs came. I wasn’t touched; many people I respect were. It was humbling and I felt lucky. That also brutally reminded me that the control was never real.

All that above marked a special 6 year anniversary.