Culture isn't what you write down, it's who you let in.

Founders often talk about culture as though it's something they can define in a deck, print onto a wall, or drop into a values page on the website and have it become real.

It absolutely, 100%, isn't

Culture is grown by the people you hire, what you tolerate, what you reward, and what you shut down early.

In a start-up, that's amplified because the team is small and every person has an outsized effect. One strong personality can lift the whole room, one toxic one can poison it.

This matters across the whole organisation, but I think it's especially important in engineering because engineering culture doesn't just affect morale, it affects delivery, product quality, decision making, technical debt, and even whether the team feels safe enough to tell the truth when something has gone wrong.

Your first hires create gravity

The first few people you hire don't just do the work, they set the tone for how that work gets done.

They become the examples that the next hires copy. They shape what "good" looks like. They influence whether your company becomes collaborative, curious and pragmatic or political, insecure and ego-driven.

That first engineering hire matters more than most founders realise.

Why? Because in many start-ups, that person is highly likely to become senior leadership later. If the company grows, today's first engineer may become tomorrow's engineering manager, head of engineering, VP, or CTO in all but title. Even if they don't formally move into leadership, they'll still become a cultural reference point.

So when you make that first hire, you're not just hiring someone to write code. You're potentially hiring someone who will teach every engineer that comes after them how to behave.

That's a huge decision.

Never hire brilliant jerks

I'll be blunt about this one.

It should be policy not to hire elitists, cargo cultists, or otherwise general assholes, even if they are the best engineers in the world.

I don't care how impressive their CV is. I don't care if they worked at FAANG, built distributed systems at absurd scale, or can visualise entire new application architectures in their head whilst making coffee.

If they belittle juniors, sneer at other disciplines, turn every technical discussion into a status contest, or treat kindness as weakness, they are not a good hire for an early stage company.

We've all been there, we've seen with our own eyes, the elitist senior who snickers at a juniors suggestion in a meeting, or outright calls them out and shoots them down. What happens? That junior never opens their mouth again - and most likely leaves within coming months. This should never be tolerated.

What you get with that sort of person is rarely just "high standards". What you usually get is fear, gatekeeping, posturing, dogma, and a team that gradually stops speaking up.

Cargo culting is just as bad, by the way.

I have very little patience for people who blindly copy architecture, patterns, process or tooling because that's what the "grown ups" do at some famous tech company. Start-ups need engineers who can think, not engineers who cosplay scale.

In recent years, monoliths fell out of fashion, everyone jumped on the microservices bandwagon. I've interviewed hundreds of candidates for engineering roles during this time and hardly any could explain why they chose that architecture, at least not with any real logical, thought through reasoning. In at least 70% of scenarios, it's the wrong architecture, it just turns spaghetti into distributed spaghetti - perhaps the reason why the majestic monolith is returning to popularity.

If someone can't explain why a practice is right for your stage, your product, your team and your constraints, they don't understand it well enough to impose it on everybody else.

Give me a strong engineer who is thoughtful, humble, curious and decent over a world-class arsehole every single time. The former raises the team. The latter eventually breaks it. You want to be looking for people eager to learn, eager to teach because every day is a school day.

You CAN guide and cultivate it

Engineering should be a learning environment

The best engineering teams are educational environments.

Not in a patronising "mentoring programme" sense, but in the real sense that everyone is learning all the time.

That includes the person with 30 years experience.

One of the worst mistakes experienced engineers can make is believing their experience means they've got nothing left to learn from newer people. That's nonsense. No-one knows everything. Technology changes, tools change, user expectations change, and sometimes the most junior person in the room sees something more clearly and more precisely because they aren't carrying 20 years of assumptions around with them. I should point out that I find that quite rare - a good engineer with 30 years experience typically underestimates their abilities and body of knowledge rather than the other way around.

A great team is one where a senior can teach fundamentals, judgement and pragmatic thinking, whilst also learning new tools, fresh perspectives and different approaches from people earlier in their careers.

That kind of environment only happens when humility is present.

The moment engineering becomes a hierarchy of who is "allowed" to have ideas, learning stops. People optimise for being right instead of getting it right. Questions dry up. Bad decisions go unchallenged. Juniors stay quiet. Seniors calcify.

That's the beginning of decline.

Psychological safety is not softness

Let's talk about mistakes, because every engineering team makes them.

Production incidents happen. Bad assumptions happen. Migrations go wrong. A change ships that shouldn't have. A security issue slips through. A rushed decision turns into a painful one three months later.

That is normal.

What matters is whether your culture makes it safe for people to bring the mistake into visibility quickly. If someone fears humiliation, blame, political fallout or a career-limiting slap on the wrist, they will hesitate. They will hide the issue, minimise it, or try to quietly fix it alone before anyone notices.

That's how small mistakes become expensive ones.

Psychological safety isn't about being soft or lowering standards. It's about creating an environment where people can say:

"I've made a mistake."

"I think I've broken something."

"I'm out of my depth here and need help."

And say it early.

An environment where there's no such thing as a daft question. Instead, ask all the daft questions, ask them on your tech wide slack channel, and watch as your good hires swarm in with help and advice, not derision.

An environment where seniors and team leads are in-tune enough to see when someone "isn't getting it" but is afraid to ask the question, so they speak up and ask the question themselves, even though they already know the answer.

That's the sort of culture you want.

Because once a problem becomes visible, the team can deal with it, rally around it, learn from it., improve the systems, the review process, the testing, the observability or the communication that will prevent it next time.

If people don't feel safe enough to tell the truth, to ask the daft questions, you don't have a high performing engineering culture. You have a fragile one.

Leaders set the ceiling

There's a hard truth here that I think founders and leaders sometimes need to hear.

Employees usually rise to the bar of their leaders, but rarely beyond it.

If leaders are sloppy, political, dismissive, arrogant or checked out, that becomes the standard. If leaders avoid hard conversations, the team avoids them too. If leaders blame, the team hides. If leaders posture, the team works towards status instead of substance.

On the flip side, if leaders are calm under pressure, willing to get stuck in, prepared to admit when they're wrong, work hard, deliver results and are consistently helpful and respectful, that becomes the norm too.

Leaders shouldn't live in ivory towers - doling out work that is beneath them to their minions. As a leader myself I have a golden rule - I'll never ask someone in my team to do something I'm not willing and prepared to do myself.

Leadership in engineering isn't about being the loudest voice or the smartest person in the room - it's about setting the example. That means being kind and considerate without becoming vague, it means leading with authority without becoming authoritarian. It means being willing to roll your sleeves up and help when things are on fire, whilst still giving the team room to own their work. It also means holding a high bar, but doing it in a way that develops people rather than diminishes them.

People watch what leaders do far more closely than they listen to what leaders say. That's why culture is curated through behaviour, not slogans.

Hire for the culture you want to compound

In the early days of a start-up, every hire compounds.

A humble, capable engineer who teaches others, asks good questions, shares context, admits mistakes and treats people well will create positive cultural gravity. More people will behave that way because that's what success looks like in the team.

An insecure ego hire compounds too, just in the wrong direction. Suddenly collaboration gets harder. Design discussions become combative. People stop asking questions because they don't want to look stupid. New hires either adapt to the toxicity or leave. The founder wonders why the team feels heavy all the time.

That isn't accidental. That's culture doing exactly what culture does.

So be deliberate.

Make it a non-negotiable that you do not hire people who are destructive, no matter how technically gifted they appear to be. Look for engineers who combine competence with judgement, humility, curiosity and kindness. Build an environment where learning is normal, where nobody is expected to know everything, and where telling the truth about mistakes is rewarded rather than punished.

Do that early enough, and the culture starts to pull in the right direction under its own weight.

Final thought

You don't build start-up culture with mission statements. You build it one hire at a time. As much as we want to believe we're defining a culture, we're not, it's emergent and evolving from the people you hire and the behaviours observed.

The first people you bring into engineering will shape how your company thinks, learns, delivers and behaves long after the org chart gets bigger.

• Hire people others can grow around.
• Hire people who make the room smarter and safer.
• Hire people who are strong enough to lead and humble enough to keep learning.

Because the culture you create in those early hires doesn't stay small for very long. It becomes gravity.