It’s commencement season, and it’s a strange world to be graduating into.
I want to offer some perspective on what it means to graduate into the world AI is making, and whether there’s still room in it for you.
Why it feels too late
“Is it too late to build something in AI?” “Does studying computer science still make sense?” “How do I find a job, let alone chart a career, when knowledge work is ending?” Underneath these questions lies the same fear: that the game is mostly over, and the prizes that matter have already been awarded.
This fear has rarely looked better founded. AI writes 75% of new code at Google. At Anthropic, that figure is north of 90%. The on-ramps that used to count as “safe” for new graduates - first-year tracks in programming, banking, consulting, and law - no longer do. Even computer science, assumed since the 2010s to be the safest major of all, posted its steepest enrollment drop on record: down more than 8% in fall 2025.
As outcomes get more extreme, the gap between the “haves” and “have nots” widens, and more people feel left behind. Eavesdrop in SF, and you’ll hear talk of a “permanent underclass.”
Alongside economic worries are others about meaning, and about what happens when we outsource too much of our cognitive labor to machines. The writing is the thinking, the debugging is how you understand the system. When we hand these moments to a model, we lose something in the process.
This is a lot to hold at once, and it lands just as life already feels like a leap off a cliff: out of the steady cadence of quarters and exams, and into a chapter of life with no preset milestones.
Why you’re earlier than you think
The first thing I want to say, as an optimist who has lived through many technology cycles: it’s going to be okay. The world is changing fast, but knowledge work is not going away.
The clearest sign is that the teams using AI most heavily aren’t shrinking: they’re hiring, aggressively. Automating the routine parts of jobs is creating more work for humans, not less. AI can complete tasks, but it can’t decide which tasks are worth doing, how they fit together, and whether the output is any good. The goalposts keep moving, and, as AI improves, human judgment, taste, and direction will continue to stay out front.
The single most important thing you can do is get on the train: start using and experimenting with AI tools every day. Approach them with a spirit of play, and see how far you can push them.
Also remember that uncertainty has always been the norm. The early-career ladder is certainly under pressure. But, for the vast majority of people, the idea that you can map out your life’s work at age 22 is a fiction. When I graduated from IIT, I didn’t know what venture capital was. I had never met a tech entrepreneur or imagined a career in software.
The same is true for you, and probably more so. Most of the jobs you’ll hold in 5 years, let alone 10 or 20, don’t exist yet. Even people in stable professions today will have to keep “firing” themselves and reinventing what they do.
I’ve written before about the difference between little-p purpose and big-P Purpose. Rather than a grand, all-consuming mission (big-P Purpose), purpose often results from small, repeated moves toward what interests you (acts of little-p purpose) that add up over time to something much bigger. Purpose is often a verb before it’s a noun.
To take the pressure off, approach your career as a prototype: something you build in order to learn from, not something you have to get right on the first try. The best teacher is contact with reality and the feedback only it can give.
What follows are a few more principles for making your way in a moment this strange and this full of opportunity.
Opportunity is everywhere. The best time to look for it is now.
In a technology cycle moving as fast as this one, it’s easy to feel that you’ve shown up too late. Having lived through several of them, I can attest that the frontier always looks claimed to those standing near it.
Consider what the world looked like to the class of 2002. The dot-com bust had just happened. The companies that mattered (Yahoo, AOL, eBay, Amazon) looked like they’d absorbed all the available oxygen. The smart move, you were told, was to go to business school or get a “real job” in finance.
Now look at the world these graduates were really entering into. Facebook was founded in 2004, YouTube in 2005, and Twitter in 2006. The iPhone debuted in 2007. Airbnb started in 2008, Uber in 2009, Instagram in 2010, and Stripe and Databricks soon after. Many of the companies and products that defined the internet for the next two decades were built by people who, in 2002, were told they were too late.
Twenty years from now, people will look back on this stretch of AI the way we look back on dial-up. You are not too late.
Being early, though, is only one part of it. The other part is training yourself to see opportunities, because they’re everywhere once you learn to spot them. Many hide inside a problem, a point of friction, or a complaint, including your own. Dissatisfaction is one of the fundamental drivers of human progress: the conviction that something slow or clunky or expensive could be 100x better.
Far from a fixed pie, opportunity is an abundant resource, which is why being genuinely happy for and supportive of your peers pays outsized dividends.
Take asymmetric risk and think in probabilities
Over the course of a career, risk and return are tightly correlated. Consistently choosing the less certain, higher-upside path is one of the biggest determinants of how far you’ll go. Loss aversion is a powerful force, but playing it safe is itself a bet, and usually a worse one than it looks.
Seek out convexity: risks whose downside is capped, and whose upside is potentially infinite. Joining a startup that folds can teach you more than it costs. Sending a thoughtful email to someone you admire invites nothing worse than silence. This asymmetry tilts furthest in your favor now, when you have the least to lose, the most time to recover, and the longest runway for good decisions to pay off.
Taking risks well also means thinking in probabilities. Almost nothing important in your life will be decided by a single choice. Outcomes are the product of many small bets, most of which you can shape in some way, but none of which you can guarantee. Your job is to maximize the odds of the outcome you want, knowing the outcome itself isn’t yours to control.
Thinking in probabilities also reframes luck. Luck is a collision: an unplanned moment when a person, idea, or opportunity crosses your path. You can engineer far more of these collisions than you think. Work for people who attract talent, share what you’re learning to draw in like-minded people, be useful before you’re asked, and follow up after you’ve been ignored.
Every smart risk you take widens the surface area of your luck.
Get insights directly to your inbox.
Chart your own course
By the time you graduate, you’ve spent two decades being measured against the expectations of others. If you haven’t already, now is the moment to stop running their race and start running your own.
I’ve watched many people who never stopped to figure out what they truly wanted end up miserable in ways they couldn’t quite name. It turns out we’re bad at predicting what will make us happy, and worse still when the yardstick is how we compare to others.
As an Indian immigrant, I know the weight of familial expectations well: the pressure to become a doctor or engineer whose success seems guaranteed from the start. I’m sure my teenage boys feel it too. And yet, the biggest lesson of my life is that, at its best, America gives all of us the rare freedom to define success on our own terms.
Doing so starts with getting clear on what you want, which is harder than it sounds. Most people don’t have a grand, world-changing dream at 22. For some, it’s earning enough to build security for themselves and those they love. For others, it’s as abstract as wanting to make the world better.
Whatever your dream is, the one thing not to do is wait on the sidelines. Moving is the surest way to discover what you excel at. Focus there, far more than on where you fall short. Leaning into a strength will carry you much further than dragging a weakness up to average.
If you can’t name your strengths, search for the through-lines in your life so far: the parts of your day that energize you; the skills you pick up faster than your peers; the projects you pour far more into than they require, simply because you love the work. Also ask the people who know you: they can often see your strengths more clearly, because to you it just feels like being yourself. Tom Rath’s book Strengths Finder is another helpful resource.
As you discover what you do best and learn to trust it, start reshaping your work around it. On small teams, people often define their own jobs. Even inside a large company, your role is, to a surprising degree, what you make of it. Give yourself permission to design your dream role.
Solve problems worth solving
Chasing trends can feel like the smart move. But while the tools change by the week, the problems worth solving change at human pace. Plenty of first-time founders start with a shiny technology and go hunting for something it can fix. The ones who build something enduring do the reverse: they start with a problem they understand deeply and use all the tools at their disposal, old as well as new, to solve it.
The leverage now available to everyone is massive. We’re entering an era where a ten-person company could be worth $10B, where the ratio between the size of a team and the size of its impact has shifted decisively in favor of the individual. This isn’t just an argument for founding something. The same leverage is yours inside a big company, on a small team, or working alone at night.
The obvious thing to do is spend this leverage on volume: ship sloppy code, write spammy emails, token-max for the sake of token-maxing. This confuses speed with velocity. Speed is pure motion; velocity is motion with direction. To win, you need to run fast in the right direction.
The more interesting thing to spend this leverage on is ambition: to take on projects you’d otherwise have lacked the resources, expertise, or confidence to start.
But ambition too needs a direction. Focus yours, above all else, on a problem that matters to you. Pick one whose resolution would mean something, both to you and a few people you know - ideally many. Then join a team or start a project that lets you go after it.
Nurture your attention and your relationships
The investments that decide a career appreciate slowly and pay off years down the line. Two compound more than the rest: your attention and your relationships.
Attention is your most valuable asset, and almost everything around you is designed to capture it. Guard it fiercely. Stay on one hard problem long enough, and your understanding compounds into mastery. That mastery is more of a superpower than ever, precisely because most people’s attention is so frayed and so scarce.
The race between Anthropic and OpenAI over the last two years is a case study. Anthropic pointed the bulk of its resources at one hard, consequential problem: coding. The depth that focus produced has outpaced OpenAI’s throw-everything-at-the-wall approach. Concentration often beats dispersion, both for a company and for a career.
Like attention, relationships also compound. Help even when you can’t see the immediate return, and don’t be transactional. The network you’ll have at 35 is built from small kindnesses you assumed no one noticed. Those who give freely end up at the center of more - more opportunities, more trust, more luck - than those who keep score.
Do the work of becoming yourself
To close, I’ll borrow from one of my favorite commencement speeches: Steve Jobs’ Stanford address in 2005.
Your time is limited, so don’t waste it living someone else’s life. Don’t be trapped by dogma, which is living with the results of other people’s thinking. Don’t let the noise of others’ opinions drown out your own inner voice. And most important, have the courage to follow your heart and intuition. They somehow already know what you truly want to become. Everything else is secondary.
The first job you have doesn’t have to be perfect. What matters is that you’re experimenting, pushing your limits, discovering your strengths, and taking asymmetric risks. The world is rich with opportunity. What I hope for you is that, when you look back, you recognize the life you built as truly your own.