One of the phrases I use a lot when talking about startup teams is: be an athlete.

I do not mean this in the “work harder than everyone else” LinkedIn sense. I do not mean grinding endlessly, ignoring your family, or pretending burnout is a badge of honor.

I mean something much more specific.

Athletes want to win. Athletes know how to lose. Athletes get coached. Athletes study the game. Athletes move on quickly. Athletes care deeply about getting better.

That mindset has always mattered in startups. In 2026, I think it matters even more.

Athletes want to win

The best athletes have many motivations. They love the craft. They love the team. They love the process. They love improving.

But underneath all of that, they want to win.

That does not mean they are selfish. In fact, the best athletes understand that winning is usually a team outcome. They know when to take the shot, when to pass, when to defend, when to listen, and when to lead.

But the internal desire is there.

They are not just participating. They are competing.

That distinction matters a lot in startups. At an early-stage company, there are always more problems than people. There is never enough time, never enough context, never enough certainty. You are constantly deciding what matters, what does not, what to ignore, and what to push through.

In that environment, I want to work with people who care about the outcome. People who feel ownership over whether the company wins. At Clarasight, we call this “be a driver not a passenger.” People who are not just completing tickets, attending meetings, or waiting to be told what to do next.

I want people who ask: What are we trying to win? What does the customer need? What is the fastest path to learning? What can I do to help the team get there?

That is what I mean by being an athlete.

Athletes know how to lose

The other side of wanting to win is knowing how to lose. This is the part I think many people struggle with.

There are a lot of people who are terrified of being wrong. They see failure as a catastrophic outcome. Not just major failure, but small failure: making the wrong call, shipping the wrong thing, asking the wrong question, proposing an idea that does not land.

Maybe this is a cultural thing. Maybe it is a social media thing, where we mostly see the polished, maximized versions of people’s lives. Maybe it is school or corporate environments rewarding certainty over learning.

Whatever the cause, it is dangerous in startups. Because startups are mostly losing.

You try a positioning angle and it does not land. You build a feature and usage is lower than expected. You make a technical bet and realize later it was the wrong abstraction. You hire someone who is talented but not right for the stage. You spend two weeks on something that teaches you what not to do.

That is not a bug in the system. That is the system.

The 2001 Seattle Mariners, the best baseball team by regular-season winning percentage in modern history, still lost almost 30% of its games. A .300 hitter in baseball is considered excellent while getting a hit only about 30% of the time. One of the greatest shooters in basketball history, Steph Curry, misses about 60% of his 3-point attempts.

That is sports. That is startups.

Failure is not the opposite of success. Failure is part of the path. The important question is whether you learn quickly enough.

The best teams have a short memory

Kobe Bryant once said: “The greatest athletes have the shortest memory.” I love that concept. It does not mean you ignore mistakes. It means you do not let mistakes become your identity.

You miss the shot, learn from it, and get back on defense. You lose the game, watch the film, and get ready for the next one. You make the wrong decision, understand why, and make a better decision next time.

Startup teams need the same muscle.

A bad product bet should not destroy confidence. A production incident should not lead to blame theater. A missed deadline should not turn into weeks of organizational scar tissue. A wrong technical decision should not become a referendum on someone’s ability.

The right response is: what happened, what did we learn, what changes now?

Then move.

The worst startup teams are both slow and fragile. They take a long time to make decisions, and then when those decisions are wrong, they take even longer to recover.

The best startup teams are fast and resilient. They make the best call they can with the information they have, they pay attention to reality, and they adjust.

Product-market fit is not a trophy you put on the shelf

This matters even more now because product-market fit is more fragile than it used to be.

Markets are changing quickly. Customer expectations are changing quickly. AI is changing what software can do, how fast teams can build, and what customers expect from products.

A company can have something that looks like product-market fit and then find itself in a different market six months later. That does not mean strategy is pointless. It means the ability to learn is part of the strategy.

In 2026, I think startup teams need a kind of curious confidence.

Confidence to make decisions. Curiosity to keep testing them. Confidence to ship. Curiosity to listen. Confidence to have a point of view. Curiosity to admit when the world is telling you something different.

That balance is hard. Too much confidence becomes arrogance. Too much curiosity becomes indecision.

Athletes live in the middle. They believe they can win, but they still watch the film.

AI raises the ceiling and lowers the floor

AI has made this dynamic even more extreme.

For great engineers, designers, product managers, and operators, AI is a force multiplier. It helps them explore ideas faster, generate options, write code, review tradeoffs, summarize context, and iterate at a pace that would have been hard to imagine a few years ago.

But AI also creates a trap.

It can make people feel productive while producing work that is not actually good.

You can generate a lot of code without understanding the system.
You can create a lot of documents without clarifying the strategy.
You can produce a lot of designs without solving the customer problem.
You can move faster in the wrong direction.

This is where the athlete mindset matters.

Great athletes know what great looks like because they have put in the reps. They have practiced, competed, lost, been coached, and developed a feel for the game. They can tell the difference between a good shot and a lucky one, between real improvement and empty motion, between effort and progress.

The same is true for startup teams.

The best engineers and product people can use AI to move faster because they have already built taste and judgment. They know what success looks like. They know what quality looks like. They know when something is elegant, fragile, overbuilt, under thought, or simply not solving the real problem.

Without that judgment, AI can hide weakness. It can generate the appearance of momentum without the substance of progress.

That is dangerous for startups, because startups do not die only from moving too slowly. They also die from moving quickly on things that do not matter.

So the point is not “use AI” or “don’t use AI.”

The point is: build teams of people with enough ownership, craft, and competitive drive to use AI well.

AI gives athletes better equipment. It does not replace the athlete.

What this means for engineering teams

For engineering teams, “be an athlete” is not about working longer hours or treating software like a contact sport. It is about how people approach the work. The best startup engineers care about the outcome, not just the implementation. They understand that the goal is not to produce code. The goal is to help the company learn faster, serve customers better, and create leverage for the business.

That requires a particular mix of traits. You need enough urgency to ship, enough humility to learn, enough judgment to know when quality matters deeply, and enough resilience to recover when something goes wrong. You need people who can make decisions with incomplete information, but who are not so attached to those decisions that they ignore new evidence.

In practice, that means building a team that can:

• Move quickly without becoming sloppy
• Hold high standards without becoming precious
• Use AI aggressively without outsourcing judgment
• Treat incidents, missed bets, and wrong decisions as learning loops
• Care about business outcomes as much as technical elegance
• Give and receive coaching without turning feedback into ego

This is the difference between activity and progress. A team can close a lot of tickets, write a lot of code, and generate a lot of AI-assisted output without actually moving the company forward. Athletic teams are different. They are always asking whether the work is helping them win.

That is the standard I want for engineering teams in a startup. Not perfection. Not heroics. Not performative intensity. Just a group of high-agency people who care about the scoreboard, learn from the film, and come back sharper every week.

The culture I want to build

The culture I want to build is not one where everyone pretends to be perfect.

It is one where people care.

Care about customers. Care about craft. Care about teammates. Care about outcomes. Care about getting better.

I want a team that wants to win badly enough to be honest about what is not working.

A team that can lose without falling apart.

A team that can be wrong without getting defensive.

A team that can move fast without becoming sloppy.

A team that can use AI without confusing output for progress.

A team that has high standards, short memory, and deep ownership.

That is what I mean by being an athlete.

And in 2026, I think that is one of the most important traits a startup team can have.